It is currently 21 Nov 2018, 09:01




 Page 21 of 22 [ 428 posts ]  Go to page Previous  1 ... 18, 19, 20, 21, 22  Next
Author Message
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 22 May 2018, 05:01 
Honest fan
Honest fan
User avatar

Joined: 11 Sep 2017, 14:24
Posts: 82
Location: Sweden
Has thanked: 1 time
Been thanked: 34 times
Quote:
But as I originally posted early on in this topic, what is the end game here??


I think that's a difficult one to answer as there are different aims (so I can only speak for myself) - The Domesday Duplicator was designed with the sole intent of making a copy of the BBC Domesday 1986 AIV LaserDiscs at a high enough accuracy so that (in the future) new discs could be produced from the copy. For my purposes ld-decode was a method to verify that the copies were good.

Once you have a virtual disc, you can load it in a virtual player - and the player can be improved over time (provided it works from the original RF sample). At the moment ld-decode provides a non-realtime conversion and isn't really recognisable as 'a virtual laserdisc player' yet - but that's what it is - and it will become more advanced.

As of where this will all lead, well, that depends on the 'glorious few' who will see a need, join in and help with the development efforts - and the end-game will be set by those (based on their wants) rather than the majority who will simply wait for something close to their needs to be produced by others.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 22 May 2018, 05:39 
Hardcore fan
Hardcore fan
User avatar

Joined: 18 Apr 2012, 18:02
Posts: 1467
Location: United States
Has thanked: 26 times
Been thanked: 19 times
@rein-o - largely preservation for me as well - it's been an interesting hobby for me, and I've learned a good bit about DSP's and a bit about electronics. It's also something that blends the past and present... we've come a long way!

@simoni - plugging my V2800 straight into the 'scope, I get about:
- 900-920mV at the end of AVS
- 400-432mV on GGV1049 #115 (the SM's reference point).
- 656-696 at the end of GGV1049
- 480-514 on a tuning fork clv disk
_________________
Happycube Labs: Where the past is being re-made, today. [meep!]
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 22 May 2018, 05:47 
Jedi Knight
Jedi Knight
User avatar

Joined: 03 May 2004, 19:05
Posts: 4898
Location: United States
Has thanked: 23 times
Been thanked: 52 times
simoni wrote:
rather than the majority who will simply wait for something close to their needs to be produced by others.

Well that will be me as I'm not program savvy and just don't see anything yet.

So I'll have to wait and see what will come of this.

But good job so far I guess, at least you have over 48 thousand who look at this topic on a regular basis.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 22 May 2018, 05:53 
Honest fan
Honest fan
User avatar

Joined: 11 Sep 2017, 14:24
Posts: 82
Location: Sweden
Has thanked: 1 time
Been thanked: 34 times
Quote:
@simoni - plugging my V2800 straight into the 'scope, I get about:
- 900-920mV at the end of AVS
- 400-432mV on GGV1049 #115 (the SM's reference point).
- 656-696 at the end of GGV1049
- 480-514 on a tuning fork clv dis


You can't really measure the RF output that way :) If you want to get a scope reading you need to add a 50Ohm load to the scope-end of your probe (I simply used a BNC T adaptor and made a BNC plug with a 50Ohm resistor on it). Some scopes (Keysight X2000s?) have a built in 50Ohm termination that you can turn on too. If you don't terminate the probe, the scope will not give the correct result - this is a common 'gotcha' and there are a number of sites that go over how to get reading of video signals correctly.

This is why I was talking about measuring it from pin 3 of the OPA690 on the duplicator board. The board has 100Ohm termination (to prevent the duplicator interfering with the player) - and it's the signal strength at that point which gets amplified by the duplicator (saves having to do boring maths :) )

Attachment:
BNC Adaptor.jpg
BNC Adaptor.jpg [ 80.04 KiB | Viewed 1247 times ]


Edit: Let me explain that a bit better :) The 100 Ohm impedance of the duplicator lowers the RF energy absorbed by the duplicator - at 50 Ohms the reduction in the overall RF energy in the player can degrade the picture visibly - this doesn't affect the duplicator or the resulting sample - but it means that the composite output of the player is degraded which I thought would confuse people. So I made the impedance of the duplicator higher (which means the OPA690 has to work a little harder) - but, in my tests, there was no visible reduction in the quality of the composite output from the player. Therefore the amplification settings of the duplicator's RF front-end has to be set with this in mind - which you can calculate, but measuring the actual signal is usually a lot faster!
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 22 May 2018, 06:05 
Honest fan
Honest fan
User avatar

Joined: 11 Sep 2017, 14:24
Posts: 82
Location: Sweden
Has thanked: 1 time
Been thanked: 34 times
Quote:
But good job so far I guess, at least you have over 48 thousand who look at this topic on a regular basis.


I realise there is a lot of interest in this work and it's great that so many people think it's a cool thing :)

There is a wide distinction between commercial projects and open-source/hardware projects - open projects are not produced to profit from the needs of others, but to fulfil the needs of those writing them - as happycube says, 'it's interesting' :)
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 23 May 2018, 17:27 
Shows curiousity
Shows curiousity
User avatar

Joined: 18 Jul 2016, 05:18
Posts: 26
Location: Australia
Has thanked: 0 time
Been thanked: 7 times
Thanks for the test card samples guys. I've made some more progress with this over the last week or so. Although I only have very basic colour decoding in place, with no fancy comb filtering techniques, I'm now getting reasonable images:
Image

Phase locking the colour burst is still the biggest problem. Getting that precise without making assumptions about the input signal is challenging, and as you can see from a colour bar test, it's still far from perfect:
Image

I'm going to keep hammering away at this, but I believe capturing at a higher sample rate would help a lot.

rein-o wrote:
But as I originally posted early on in this topic, what is the end game here??
Will there be some type of decoder for people to buy, is this going to get us regular LD collectors close to getting a new working player or is this just a hobby
like that Radio Shack thing I had when I was a kid to learn electronics??

Different people will have different ideas and motivations. Personally, I'm less interested in the decoding of raw RF from laserdisks. What interests me is getting general purpose composite video decoding solid, and realtime. Basically I want a software-based analog video capture solution. It'd be nice to shrink it down onto a simple system like a Raspberry Pi (although the current one can't do it) which can make a perfect composite to HDMI converter. Every one I've ever tried has been garbage. I'd also like to be able to get good full frame captures of any analog video stream on PC. Nothing in the market can do it right now it seems. I'm also interested in the Domesday Duplicator hardware itself as a general purpose data acquisition device, for sampling any number of analog signals. I do a lot of work reverse engineering old hardware, and a capture device like this is useful for capturing large amounts of raw data for analysis. I'm also interested in capturing and decoding raw CD RF signals, as there's a lot of potential to improve archive processes there. All current CD image formats are incomplete and flawed, and are based around what can be read on a traditional PC CDROM drive. Hell, nothing can even store the true TOC data from the lead-in as part of the image, and very few formats even support scrambled sectors. I've already got a problem with the digital data I've ripped from the "MegaLD" disks on the Pioneer LaserActive, in that there's no CD format in existence that can hold all the data, I have to store it in my own custom way. The best encoding is clear: straight digital data in the same format it's stored in the disk, with all the original error codes still in place. It's easy in that format too to rip the same disk 3+ times, and combine the datastreams to eliminate per-disk errors, leaving you with a true 100% correct duplicate of the original master. This is sorely needed, and the work already done here puts us only a stone's throw away from this goal.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 23 May 2018, 17:36 
Shows curiousity
Shows curiousity
User avatar

Joined: 18 Jul 2016, 05:18
Posts: 26
Location: Australia
Has thanked: 0 time
Been thanked: 7 times
Since I've brought up capturing composite video over the Domesday Duplicator hardware, any thoughts about how that could be done simoni? According to my reading of this page a voltage range of -300mv to 1000mv should be enough for any signal type. How would I go about configuring the front end to handle a signal like that?
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 23 May 2018, 18:05 
Genuinely interested
Genuinely interested
User avatar

Joined: 26 May 2014, 19:25
Posts: 38
Location: United States
Has thanked: 2 times
Been thanked: 1 time
I'm curious about something that I'd like to get out of my mind and have someone like Simon or Chad comment on the possibilities...

I know there will never be a new Laserdisc produced, ever. Done. However, what about a CDV?

I'm not sure what the structure of a CDV is but it is CDDA and LD RF on a 12cm disc that can be played in both CD only players and CDV/LD players. It must be redbook compliant (or at least very similar). It also must be LD compliant.

Now, we have the technology to capture the raw RF signal of LD (thanks guys, really). We can also burn CD-R media at home. Would there be a way to burn a compliant CDV at home with LD RF information readable on a physical player? I think I've tried to copy/rip a CDV and you only get the CD part. If you could control the drive via software could I copy the whole disc as it is with the CDV part in tact? Perhaps some kind of RAW mode that reads the entire disc? The pit size and laser specs are the same right?

Obviously this would only be for a small run time and only be a novelty since you'd be limited to whatever you capture from an existing LD source, but then you could make physical "copies". I know there are a lot of anime fans here and maybe a single episode might fit on a single CDV-R...

However, what if we could encode video (say from a video camera) to RF as used in LD? Could that then get encoded and burned to a CDV at home using open source software that has yet to be written?

Is it possible? Please say yes ;) Otherwise, just say I'm crazy and I can accept that also :)
_________________
Kevin
Onkyo TX-SR705|CLD-M301|CLD-D406|CLD-D704|LX-900U|APD-1|Sony XBR55X810C
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 23 May 2018, 19:01 
Honest fan
Honest fan
User avatar

Joined: 11 Sep 2017, 14:24
Posts: 82
Location: Sweden
Has thanked: 1 time
Been thanked: 34 times
Weeee... so many questions :) Let me see what I can do...

Quote:
I'm going to keep hammering away at this, but I believe capturing at a higher sample rate would help a lot.


This can't be the case ;) Laserdisc players can do it with half the sampling speed, vis-à-vis, so can you :) I looked into this quite a bit and came up with the conclusion that the only sure-fire way was to do it is by using a ADPLL implementation (All Digital Phase Locked Loop). When I looked at the best way to achieve a constant sub-sample timing measurement (which was resistant to error and jitter), everything pointed to this. Trying to do it in a 'deterministic' fashion (like ld-decode does at the moment) will never get the consistent results that a PLL would. Trouble is, I couldn't for the life of me figure out how to code it. I even paid for a copy of Phase Locked Loops by Roland E. Best and just ended up in a flat-spin around the maths involved. DSP coding isn't really my forte! (I can do FPGA, embedded C and desktop C++ though, so I won't beat myself up over it ;) )

You seem to have a better grip of signal processing software than myself though; so perhaps my insight will give you an idea (even if I'm horribly wrong!).

Quote:
According to my reading of this page a voltage range of -300mv to 1000mv should be enough for any signal type. How would I go about configuring the front end to handle a signal like that?


First of all you have to understand the difference between the AC component and the DC component of a signal. The DC component of the signal you describe has a minimum of -300mV and a maximum of 1000mV with a centre-point (ground) of 350mV. The AC component of the signal has a peak-to-peak of 1300mV. The first thing the duplicator does is remove the DC component of the ingress signal. This puts the centre point on 0V (this is what C24 does). Then it applies an impedance (R1) which, well, impedes the signal (making it smaller, but preventing the rest of the circuit becoming the 'path of least resistance' - otherwise the duplicator would gobble-up all the available RF energy and the player would stop working - this is why video is typically either 75 or 50 ohm impedance (simply speaking)).

The next step is to add back in the DC component required by the ADC. That has a DC range of 1500mv to 3500mV. R2 and R3 divide the reference voltage and apply it to our signal. Moving it from 0mV ground to 2500mV ground. It then feeds it to the opamp which is configured as an amplifier.

The opamp's job is to get the AC component of the signal as close to the 2000mV peak-to-peak limit of the ADC as possible (to get the best use of the available 10-bit resolution) - but without changing the DC component. So the resistor settings made by R5 and R4 add gain. So in your example the gain would need to get you from 1300mV to 2000mV plus whatever you lost in impedance. The end result should be a 2000mV (max) AC signal with a DC offset of 2.5V.

The real complex stuff is getting 40 million 10-bit numbers a second from there to the PC's harddrive (and keeping the noise down); but I've already solved that for you :)

That's quite possibly the longest "yes you can" answer you will get this month! But, I hope it will help you to understand more about what you are asking.

Now sonicboom:

Quote:
Is it possible? Please say yes


Yes - but then so is making new laserdiscs. If they did it in the 70s - we can do it now. The real question is why would you want to? Well, there is 'fun' which is a very strong driving force, but there isn't 'profit' which is the strongest force known to mankind. The beauty of an RF sample is it's really a disc, just stored in a different medium. Why not just put an SD card and FPGA together and make a matchbox sized, all digital LaserDisc player... just based on the RF samples instead of real discs... Hell, you could even hook it up to the RF out of a real player and do it all in realtime... just needs a "little" development; same for CDs/CDVs/whatever :)
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 25 May 2018, 05:21 
Hardcore fan
Hardcore fan
User avatar

Joined: 18 Apr 2012, 18:02
Posts: 1467
Location: United States
Has thanked: 26 times
Been thanked: 19 times
You're right, the values are quite different with 50ohm termination on my 'scope:

- ~200mV ggv 115, ~312 ggv end
- ~408 AVS end

Looks like the attenuatior above a certain voltage lets too much of the audio signals (esp CD/EFM) get through, so I'll hopefully take it to the local Hackerspace tomorrow for a resistor swap.

Other random not-enough-coffee-and-slept-on-my-shoulder-wrong-again thoughts:

- @nemesis - looks like you're making good progress! Something that worked well for me was looking at a line and the two color bursts around it - my current TBC takes 2-3 passes.

- The DE10-Nano has an ARM SOC + FPGA, with HDMI output. With a lot of work it might be possible to do real time video decoding through that.

- I think a data stream format that reflects the EFM subchannels is going to be in order sooner or later. It would be good for [C/L]D-G. It doesn't have to be completely raw EFM but it could have a packet format that covers anything non-zero.

- CDV-R's existed in Pioneer's lab at least. They had special blanks... in theory someone could hack up a burner for it, but it would be a ton of work I doubt anyone will ever do. (but feel free to prove me wrong there!)

- If flash keeps getting smaller/cheaper we're not that many years off from affordable 256GB microSD cards that could store a disk's worth of 8-bit LD RF. Or four Blu-Ray rips. Or who knows how many hours of SD video with modern encoding ;)
_________________
Happycube Labs: Where the past is being re-made, today. [meep!]
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 26 May 2018, 17:13 
Genuinely interested
Genuinely interested
User avatar

Joined: 26 May 2014, 19:25
Posts: 38
Location: United States
Has thanked: 2 times
Been thanked: 1 time
Quote:
Is it possible? Please say yes


Quote:
Yes - but then so is making new laserdiscs. If they did it in the 70s - we can do it now. The real question is why would you want to? Well, there is 'fun' which is a very strong driving force, but there isn't 'profit' which is the strongest force known to mankind. The beauty of an RF sample is it's really a disc, just stored in a different medium. Why not just put an SD card and FPGA together and make a matchbox sized, all digital LaserDisc player... just based on the RF samples instead of real discs... Hell, you could even hook it up to the RF out of a real player and do it all in realtime... just needs a "little" development; same for CDs/CDVs/whatever :)


Why? For the novelty of being able to do it and have something physical to show it on any player, not just a computer/flash card. As you say, the fun factor. You like programing and I like burning physical discs.

My thought was that not everyone has access to test patters on LD to calibrate physical players. Maybe ripping and burning those onto CDV-R would be a way to get around that since no one is producing anymore Video Essentials discs. I lucked out and found VE locally twice for under $5. There is a pretty high demand for test discs and not much supply. Even the Pioneer discs will eventually be discontinued and you'll never see those in the wild. However, as you mention you could make a magic box that plays the decoded rip through a physical player and do the alignment that way I suppose. Then again, who's making those and for how much? Right.

I'm grateful for the Pioneer Tuning Fork and Alignment videos that were RF captured, decoded and put on You Tube. I would never have had access to those videos otherwise. A perfect example of how this is a genius idea; making rare or hard to find discs available to a wider audience at the highest possible quality. Again, not so much for movie archives. That is my thought with the test patterns on CDV-R.

LD collecting in a world headed towards streaming UHD is really just romantic nostalgic novelty anyway, based on the physical size and resolution. If LD wasn't a 12" CD with a movie on it and cool art work would anyone care these days? Probably not. So, for movies at least, an RF capture of a LD is just a Standard Definition NTSC or PAL stream like any other on Netflix or Hulu, etc. Where's the fun in that? Obviously there are several LD titles only available on LD so there's an incentive. Your initial intention is not movies per se, so I get it, but most LD enthusiasts are in it for the movies.

As cool as it is, the learning curve or investment in the LD Decode and capture board process will keep it a novelty for most LD collectors until everything is worked out. It isn't for everyone yet. Most people, I assume, will hold onto their physical players until they die and continue to watch their collections on LD not on a hard drive or flash card. Even if it was off the shelf and under $100 the storage space required to capture a single disc (500 GB) is prohibitive let alone an entire collection. This is no doubt the best way to make archival 1:1 "back-up" copies of a LD. It is genius and you're hard work is commendable.

I appreciate the energy in responding. I know between the FB page and this forum you've swamped with questions from people like me who just don't have the skill set you do ;)

Thanks
_________________
Kevin
Onkyo TX-SR705|CLD-M301|CLD-D406|CLD-D704|LX-900U|APD-1|Sony XBR55X810C
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 03 Jun 2018, 09:14 
Honest fan
Honest fan
User avatar

Joined: 11 Sep 2017, 14:24
Posts: 82
Location: Sweden
Has thanked: 1 time
Been thanked: 34 times
So a quick update from the Domesday Duplicator project... There is a new build, now under final testing, that improves the clock accuracy of the board and allows the ADC to be pushed all the way to the 40MSPS limit :) The duplicator set-up is now streaming at over 610Mbits/sec generating around 76Mbytes of data every second.

Pretty shortly there will be a new release of both the FPGA firmware and the Linux GUI. I've been working a lot with happycube analysing the resulting samples and there is some upper-sideband video at around the 16MHz point that was distorting the signal (this was missed by the original TV capture card as the low-pass filtration was probably much lower in the signal). Increasing the duplicator bandwidth to 20MHz captures everything (and this is preferable to lowering the stop-band of the LPF as that could clip the real signal); so the hope is that the end result will be a totally accurate capture.

There's a lot of detail on the Domesday86 facebook group (https://www.facebook.com/groups/2070493199906024/) showing the results of the on-going testing if anyone is interested in the nitty-gritty stuff; otherwise, watch this space for the new build :)
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 03 Jun 2018, 19:11 
Honest fan
Honest fan
User avatar

Joined: 11 Sep 2017, 14:24
Posts: 82
Location: Sweden
Has thanked: 1 time
Been thanked: 34 times
The fourth release of the Domesday Duplicator solution is now available (40MSPS sampling). All testing will be performed on this version from this point onwards, so upgrading is very much recommended:

https://github.com/simoninns/DomesdayDuplicator/releases/tag/V1.6
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 05 Jun 2018, 19:18 
Honest fan
Honest fan
User avatar

Joined: 12 Sep 2015, 05:57
Posts: 67
Location: United States
Has thanked: 1 time
Been thanked: 1 time
@reino: some of the reasons I (personally) was interested in this project, were that the LD_DECODE captures of the jason PAL test card revealed noticeably higher resolution than any players could produce, indicating possibly there was more to be had from at least PAL LDs, than off the shelf players can produce. I replicated this capturing as well, however captures of CLV discs did not produce significantly higher resolutions :( . Additionally, most of the players I have are in some degree of deterioration. The components inside degrade over time (especially capacitors). I have several models for which I own more than one player, and they all have different PQ. Capturing the signal directly and decoding in software bypasses that deficiency.

@nemisis: it was a week or more ago that I read all your posts, so forgive me if I am mistaken in my understanding, but did you say you were looking for a way to capture the composite signals? https://www.rtl-sdr.com/tag/pal/ < that article talks about broadcasting a composite signal using HACKRF type device. if the hackrf can transmit composite, it should be able to receive/capture it as well. The article talks about PAL, but could be asjusted for NTSC. Secondly, did you say you weren't happy with the composite decoding of devices you tried? I know the $39 composite-hdmi adapters on amazon or at your local big box store aren't very adept, but professional grade decoders seem to do a dang good job in my experience. I have the Blackmagic Designs teranex 2d that I got on ebay for a very reasonable price, and it seems to decode composite signals (and offer a bunch of other modifications: scaling, denoising) in real time quite impressively.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 07 Jun 2018, 13:34 
Second post
Second post
User avatar

Joined: 18 Jul 2013, 00:11
Posts: 2
Location: Japan
Has thanked: 0 time
Been thanked: 0 time
There has been some really amazing progress in my absence!

Regarding the Facebook group/discussion: "One does not simply log into Facebook..."

Maybe we could just do a thread here on the technical sub-forum? Though I appreciate I'm hardly an active contributor here.

On the EFM decoding - I made a couple of amateurish attempts to create a D-PLL, but I think my efforts were hindered by the input sample rate being too low relative to the bitrate (i.e. ~4.1Mhz) - the resulting phase-error signal was too coarse/low resolution and as a result it was impossible for my PLLs to get a stable lock on the signal.

Up-sampling the filtered signal to say 20-40x the channel bitrate (i.e. 40-80Mhz) would potentially allow the algorithm to be more stable, but I got stuck thinking about how to do this without adding a 80 megabyte-per-second re-sampling step (i.e. using some kind of bi-linear interpolation). However, as noted elsewhere, there may be some odd stuff going on with the signal at lower frequencies which might hinder this approach.

I'd be happy to try and put together a Doomsday board to have another bash at it, though I need to first clear the decks here a bit.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 07 Jun 2018, 15:53 
Honest fan
Honest fan
User avatar

Joined: 11 Sep 2017, 14:24
Posts: 82
Location: Sweden
Has thanked: 1 time
Been thanked: 34 times
Quote:
Regarding the Facebook group/discussion: "One does not simply log into Facebook..."


As that maybe, the discussion I was referring to was the Domesday86 project (the Domesday86 project consists of 3 sub-projects at the moment of which one is the duplicator and the other 2 have very little to do with LaserDiscs). I'm happy to post general updates to multiple forums, but the details (on the Duplicator) will be via the Github issue tracker and the Domesday86 Facebook group as my time is limited. This thread is really about ld-decode (which, I would guess, is the technical bit you're interested in) rather than the electronics-detail of the duplicator board; so I'm sure happycube will report progress here (as will I as new releases become available).

Quote:
Up-sampling the filtered signal to say 20-40x the channel bitrate (i.e. 40-80Mhz) would potentially allow the algorithm to be more stable, but I got stuck thinking about how to do this without adding a 80 megabyte-per-second re-sampling step (i.e. using some kind of bi-linear interpolation). However, as noted elsewhere, there may be some odd stuff going on with the signal at lower frequencies which might hinder this approach.


You don't add more information to the signal by interpolating to higher-frequencies; so if what you need is in the upsampled signal - it's in the original too :) Having said that, the Duplicator samples contain a lot more information, so it should be a more 'doable' approach (and are already running at 40MSPS). There do seem to be ways to get sub-sample resolution from a ADPLL; but I wasn't personally able to get my head around the maths...
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 08 Jun 2018, 13:40 
Second post
Second post
User avatar

Joined: 18 Jul 2013, 00:11
Posts: 2
Location: Japan
Has thanked: 0 time
Been thanked: 0 time
Cool.

HappyCube sent through a Doomsday sample of a CD. It looks promising - specifically the edge transitions are extremely crisp.

Quote:
You don't add more information to the signal by interpolating to higher-frequencies


It was more about feeding a (shoddily implemented) algorithm intended for a much higher samples-to-signal ratio an appropriate rate of samples. Having more samples to start with is a much better place to be.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 08 Jun 2018, 13:59 
Honest fan
Honest fan
User avatar

Joined: 11 Sep 2017, 14:24
Posts: 82
Location: Sweden
Has thanked: 1 time
Been thanked: 34 times
The Domesday Duplicator is functioning well; but there are some issues with the RF front-end. It's good, but it's not good enough yet. Well, let me qualify, it will make an excellent copy of a LaserDisc (or a CD I would think); but for the Domesday86 project, I need near-perfect copies... So, with happycube's help (he is testing and verifying decoding and analysing what causes issues), I'm redesigning it, so the next hardware version will be even better.

By matching the filter design of the duplicator with the filters in ld-decode it should be possible to pull-back the original signal as the end-to-end chain of processing is known. The closer this match gets, the better the resulting decode will be.

Hardware is just like software in that it requires testing and iteration too; so the slow uptake of people building boards is beginning to show results; the more samples there are, the better the design can be. My plan is for the next hardware revision to be good-enough for mainstream use. However, one can never be 100% certain :)

Interestingly; I've actually tested the duplicator with an ADS828 chip now - it happily performs at 75 million samples/sec... massively more than required for the application, but nice to know there is some headroom if needed.

Footnote: It's Domesday rather than Doomsday - the word is old English, means "Day of Judgement" - nothing to do with 'doom' :) ). There's a light introduction here if you'd like to read it: https://www.domesday86.com/?page_id=2140
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 20 Jun 2018, 18:58 
Honest fan
Honest fan
User avatar

Joined: 11 Sep 2017, 14:24
Posts: 82
Location: Sweden
Has thanked: 1 time
Been thanked: 34 times
The new 3_0 release of the Domesday Duplicator is now available: https://www.domesday86.com/?page_id=2233

This new release contains a redesigned 3_0 PCB providing three major new features:

1. Elliptic 2nd order filter – greatly reduced aliasing and better overall gain performance across the entire sampling bandwidth
2. Configurable gain – Design now includes a user-configurable gain setting allowing the board to work with the majority of LaserDisc player models (once adapted for direct RF output)
3. 50 Ohm impedance – the Duplicator board is now impedance matched for use with 50 Ohm RF output providing better, predictable performance

In addition the board layout has been improved so component identifiers now include the schematic sheet number, so it’s much clearer what components are performing the various board functions.

Attachment:
The Domesday Duplicator 3_0 small.jpg
The Domesday Duplicator 3_0 small.jpg [ 67.58 KiB | Viewed 755 times ]


It should be noted that the samples generated by the new 3_0 board are not ‘compatible’ with the 2_x versions of the PCB (due to the new filter configuration). 2_x boards are now considered obsolete and should be upgraded to 3_0 (the majority of the components are reusable on the new board).
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 28 Jun 2018, 13:50 
Shows curiousity
Shows curiousity
User avatar

Joined: 18 Jul 2016, 05:18
Posts: 26
Location: Australia
Has thanked: 0 time
Been thanked: 7 times
Wow, this looks like a huge step forward. I'm really excited about the configurable gain options in particular, and that'll make it much easier to use this hardware to capture a wider range of signals.

I know this is outside the scope of what you designed this for, but what are your thoughts simoni on using this setup as a logic analyzer? I'm planning at some point soon to have a crack at bypassing the ADC, and pumping digital signals straight into the FPGA. I might even try my hand at prototyping a new board based around this usage. I'm assuming it should be relatively easy to achieve 16-bit digital capture on your current design, but I'm hoping to add modes that would allow up to 48-bit capture, obviously with a lower maximum sampling rate to account for the increased data size. Do you have any thoughts on this, such as how easy it might be to achieve, or potential barriers to using the hardware in this manner?
Offline
 Profile  
 
Display posts from previous:  Sort by  
 Page 21 of 22 [ 428 posts ]  Go to page Previous  1 ... 18, 19, 20, 21, 22  Next


Who is online

Users browsing this forum: No registered users and 8 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

Jump to: