AVS Forum banner

Status
Not open for further replies.
1 - 19 of 19 Posts

·
Registered
Joined
·
258 Posts
Discussion Starter · #1 ·
I brought this up over in the videoprocessing forum, since they seem gung-ho about feeding the digital output from a stand-alone DVD player directly into D-scaler via a $500 SILK card. From what I can gather, horizontal pans are smoother this way.


Nevertheless, I'm surprised there isn't a way, somehow, to feed the video data directly into dscaler from a DVD-ROM drive instead ?!?


This apparent obsession they have concerns me about the limitations of 100 % software based DVD players, despite the conventional wisdom on the board.


Can hardware-assist can eliminate jittery horizontal pans ? How smooth is the Hollywood + card hardware decoder ?


CG
 

·
Registered
Joined
·
3,316 Posts
DScaler interfaces directly with the hardware of the BT chipset and there is no easy way to interface that with the DVD drive. You could do something goofy with SDI out from HTPC back into the SDI Silk card, but then you lose the whole reason you got SDI in the first place, no judder and easy interface.


This sort of stuff has been coming up lately with people making weird requests of DScaler...
 

·
Registered
Joined
·
67 Posts
Hi Fredzy,


I share the same interest as you on this question. Unfortunately there is a limitation involved in doing this. Here's what I gather from the SDISilk FAQ:

Will there be any way to get the SDI advantage with the DVD player in your home theater computer?
We don't know. The DScaler developers have looked at this a lot, but there appear to be challenges to making it happen at a satisfactory quality level for now.


As far as the Hollywood Plus card, I don't really recall it having juddery pans but it's been years since I've used that card. As long as you had the correct refresh rates it should not be a problem. Hopefully someone with more extensive knowledge will jump in and give a better answer concerning the dvdrom limitations.
 

·
Registered
Joined
·
13,423 Posts
â€This apparent obsession they have concerns me about the limitations of 100 % software based DVD players, despite the conventional wisdom on the board.


Can hardware-assist can eliminate jittery horizontal pans ? How smooth is the Hollywood + card hardware decoder ?â€



CG,


I’ve recently discovered an awesome hardware decoder/scaler for DVD. The MyHD card can decode/scale non-CSS encrypted DVDs and the image quality and smooth motion is great.
 

·
Registered
Joined
·
1,804 Posts
Any plans for making it possible to playback actual DVD's on the MyHD card? I mean, 99% of all DVD's I want to watch would be encrypted...


As I suggested for the Hipix guys, I would pay for a software solution that could do that (since I would not have to switch my output cable from the HDTV card to the VGA card every time I want to watch a dvd...)


The Hipix team seems to have even decoding of unencrypted dvd's very low on the list of things to do, so I am considering ditching the telemann card and go for the MyHD instead... (PS: I do know about the independent initiative that one guy made to converte vob->ts that hipix can play, but it still has problems, on no-one seems to want to/be able to help him)


-Tom
 

·
Registered
Joined
·
13,423 Posts
â€Any plans for making it possible to playback actual DVD's on the MyHD card? I mean, 99% of all DVD's I want to watch would be encrypted...â€


Tom,


When they do provide “actually†DVD playback it will be restricted to 480p on the MyHD due to licensing requirements.
 

·
Registered
Joined
·
3,316 Posts
Cliff,


Since you aren't part of the company, maybe you can answer this. What do you think the chance of there being a registry setting to enable higher than 480p playback? Kazaa does this with 128kbps MP3s to stay legal with European law I believe...their software comes perfectly legal and if people decide to break the law, then the company isn't held responsible.
 

·
Registered
Joined
·
13,423 Posts
Cal,


I don’t believe MIT is that stupid. The company that exposed the CSS decryption key and made programs like Smartripper possible is no longer in business.
 

·
Registered
Joined
·
1,804 Posts
Cliff, does anyone know why the DVD licensing would restrict anyone to output a max of 480P?? I don't understand why this could possibly be an issue. Shouldn't they even outlaw programs like WinDVD and PowerDVD, which you can maximize in higher resolutions than that? Or what about all graphic card vendors, for making hardware that can be misused to show DVD higher than allowed resolution? Also the SDI Silk boards should definitely be outlawed then, since people in this forum obviously use this to "break the law"


I guess I have a problem of seeing why getting better picture (from an identical source as everyone else) would be illegal...


And also, as you might guess, why ban one source, but not others..


Hmmm..rambling on here, just curious :)


Ps:thanks for all the help you are offering to this community, Cliff, I have definitely learnt a lot from you, as I am sure many others have as well.
 

·
Registered
Joined
·
13,423 Posts
Tomes,


Don’t get me started. My blood pressure can’t stand it. ;)


Firstly there is NO REAL law restricting the resolution to 480. It is an attempt to keep MicroVision from going out of business because technology has made them obsolete and as MV derives their income thru the DVD Corksockers Group that license DVD decryption with MV. They are using this restriction to force manufacturers to use MV thru their licensing dicked-ums.
 

·
Registered
Joined
·
9,884 Posts
Cliff -


Tell us how you really feel. ;)


I hesitate to jump in here after that, but obviously in my case you are already preaching to the converted. And I'm going to speculate a little.


I can't prove it but I think that DVD licensing is limited to 480 lines (what about PAL?) only if they are being displayed on a TV. Computer display on RGB is still not completely locked up. But any hardware that can display to s-video or component analog is assumed to be displaying to a TV and subject to the rules.


I think that is why Stephen Orr suggested that all ATI video cards would be required to have TV out in the future, or some such silly rule.


But for now if there is not an active TV out and you just use a VGA->YPbPr transcoder I'm not sure if it is covered under those rules. The true test will be if PowerDVD and WinDVD come out with broken versions that can't display large screens.


Possibly HDTV cards could also make an end run around this by supporting only RGB computer displays and forcing the user to get a transcoder that most of us may have already.


And of course DScaler and Silk cards don't need a DVD license anyway.


I'd really love to see the writing of the actual DVD licensing agreement. There may be a case to be proved for anti-competitive practices in discriminatory licensing there if all the interlocking side effects were made known.


Besides, it is just stupid. ;)


- Tom
 

·
Registered
Joined
·
1,141 Posts
I may be missing something here, but I have had my AVS HTPC for 3 months now, and have never noticed any jittery pans, or any artifacts at all for that matter.


Exactly what is it that SDI Silk or MyHD are supposed to "fix"? I watch a ton of DVDs and haven't had a problem yet...and I consider myself very discriminating about anything on my system that's not perfect.


Am I just lucky in this regard with my setup, or are ther certain discs that show the "problems"?


I love what I have, but, like most here, am always willing to tweak for PQ.
 

·
Registered
Joined
·
3,316 Posts
Your HTPC doesn't give bad performance, just not as good as SDI as I understand it. Very big pans still look kind of bad to me on my HTPC which looks amazing most of the time.


For a worst case scenario, rent Snake Eyes with Nicholas Cage. They have a 360 degree pan around the casino that is hideous.
 

·
Registered
Joined
·
13,423 Posts
Tom,


It is stupid and I agree that it is anti-competitive practices.


RGB DACs and now component (YPbPr) DACs have never supported MV encoding. The actual MV signal is generated by the NTSC chip used to output interlaced 480 to a TV and is triggered by a bit in the digital stream coming from the DVD. I have no objection to that because an NTSC chip can’t process above 480 anyway, but I do object to the restriction of technology that does not support MV. Has anyone seen a VCR type recorder that can record analog RGB or analog component that a consumer can afford?


What difference does it make where the scaling takes place? Does it matter if the DVD is decoded to the native resolution of 720x480 and then scaled to 1080i in a PC card or in a HDTV? All of the licensed soft DVD decoders comply with the CSS/MV requirements by decoding to 480p. It is the built-in PC scalers they are trying to restrict.


It would not surprise me that some of the manufacturers that are in bed with the DVD Group goes one step beyond the 480p restriction by adding MV type trash (image interference) in their hardware scaler to force playback to 480p even on not CSS/MV DVDs. Of course if they do that the product will be DOA.
 

·
Registered
Joined
·
88 Posts
I'm working with some people on a Linux-based DVD player. By player, I mean box, not player-software.


We're still experimenting with Xine & Ogle, but the plan is to turn one of those into a frame server and hook dScaler to that (ain't open source great?).


The plan is to eventually build a better scaler, all in software, writing to the RGB framebuffer, not the YUV hardware overlay plane. That's going to require a lot of juice, so we're looking at 2GHz P4, using SSE2 in inner loops.


We want to build a box using a motherboard with built-in video (P4S333-VM), a DVD-ROM drive (Pioneer slot-load), a quiet power supply (having trouble there- itty-bitty quiet supplies like the 1U PC Power & Cooling don't have enough amps @ 12V for P4) and a Compact Flash to IDE-cable adapter.


Yep, no hard drive. It would boot right off the card.


We'd sell it as a "kit," missing libdvdcss sort of the way you can buy automatic weapons as semi-automatics, then a modification kit from the guy at the next table at the gun show. Probably ougth to come up with a better analogy, though.


libdvdcss would be OK from the DMCA point-of-view because it would be a closed box and the 480p limit is a licensing issue, not a legal one. Our problem is the trade-secret status of CSS, though by the time we're ready, that might be resolved (it's in court now).


It won't be a full fledged HTPC & it won't scale external sources (v. 2 maybe?), but it'll be a nice HD-monitor compatible DVD player.


We all used to work here and built some of the software that runs in the Scientific-Atlanta Explorers (if you're a Time-Warner digital cable customer, you're running my code) so the only hurdle is finding the time....


-M


Yes, the contributions to Xine & dScaler would be contributed back (or available as a patch if they don't want 'em) so you could run this on a HTPC you build, but at this point the plan is to make the 2nd gen scaler closed-source.
 

·
Registered
Joined
·
226 Posts
Hi Jeff,


Give the SDI Silk a try. You of all people will probably notice a better picture quality because of your setup. Check out the following thread http://www.avsforum.com/avs-vb/showt...hreadid=129075

Look for Cliff's comments about sharpness and colours that beat software players. Maybe it's the MPEG2 decoding that's better or maybe because it avoids direct show. SDI to the Vigatec when Alan first got it made him rethink his ATI based HTPC as his reference standard then. He also made comments similar to Cliff about resolution and colours. Plus, you will get better de-interlacing of video based DVDs.


If you do decide to go this route, you will probably need a recalibration of your grey scale. I might be wrong about that. If you do, you might as well get the upcomming Matrox Parhelia video card as well. Should be sharper, more accurate colours, less ghosting, less pixel jitter than ATI, which Chris Stephens is extremely concerned with to have repeatability and better resolution. You can look up Chris' comments on pixel jitter in his special guest thread.


Ben
 

·
Registered
Joined
·
2,379 Posts
morganw: Why don't you send an overview of your plans to the DScaler mailing list? I'm sure everyone would be interested in hearing what you have in mind.
 

·
Registered
Joined
·
9,884 Posts
Quote:
GB DACs and now component (YPbPr) DACs have never supported MV encoding. The actual MV signal is generated by the NTSC chip used to output interlaced 480 to a TV and is triggered by a bit in the digital stream coming from the DVD. I have no objection to that because an NTSC chip can’t process above 480 anyway, but I do object to the restriction of technology that does not support MV.
Cliff -



Sadly, I'm not sure that's true anymore. From the DVD Faq :

Quote:
There are three forms of copy protection used by DVD:


1) Videotape (analog) copying is prevented with a Macrovision 7.0 or similar circuit in every player. The general term is APS (Analog Protection System). Computer video cards with composite or s-video (Y/C) output must also use APS. Macrovision adds a rapidly modulated colorburst signal ("Colorstripe") along with pulses in the vertical blanking signal ("AGC") to the composite video and s-video outputs. This confuses the synchronization and automatic-recording-level circuitry in 95% of consumer VCRs. Unfortunately, it can degrade the picture, especially with old or nonstandard equipment. Macrovision may show up as stripes of color, distortion, rolling, black & white picture, and dark cycling. Macrovision creates severe problems for most line doublers. Macrovision is not present on analog component video output of early players, but is required for newer players such as the Sony S700 (AGC only, since there is no colorburst in a component signal).
So I think my memory (nightmare) that they had added it to component video may indeed be true.


- Tom
 
1 - 19 of 19 Posts
Status
Not open for further replies.
Top