[KBK]

I can certainly do the hardware mod if the format's are compatible.


Is the encode format of the DVD drive compatible with the format that the parallel HDTV port on a conexant 878A chip is looking for? If so, it may be possible to run the DVD drive right into the capture card, shut down any software DVD functions,and then... run Dscaler algorithms on the DVD's data...and use it instead of all the software devices that are not getting us the PQ we are looking for. I realize that the sound function is not there, but, is anyone interested, specifically the Dscaler development team?


After all, most of the software is already in place.


After all, this is the next step, you see....create our own scaler functions.


Now, is this a possible thing, or is this not worth the trouble?


I also fully realize that such a thing begins to border on the single box solution, with a multi-pronged attack for PQ problems. Too bad about the sound being a problem due to licensing. The remote fucntion would ahve to be put in place. After all, the cards all seem to come with a remote.


Too bad it will probably never happen due to the fact that it is too close to the commercial interests. Oh well.

 

[KBK]

You see, the SDI option on the new software is porting to the parallel data port onthe 878A chip, and maybe this woukld be useful in other ways...

 

[Hugh Magoo]

this sounds interesting, so I'll hold it out and if anyone knows exactly what KBK is talking about they can help him out.


help him help us.

 

[luvmytivo]

Ken,

. . . AFAIK, any current DVD-Rom drives won't actually output any video, only bits of data that are decoded by the CPU. The long term solution is probably going to be a software based convergence of DScaler algorithms with a software MPEG2 decoder. The modular direction the DScaler team has taken with regard to it's scaling algorithms should make this an eventuality. Unfortunately, an open source DVD player is a tough road to hoe thanks to the DMCA and it's implications on CSS, but it's not insurmountable.

. . . When confronted by the cost of SDI cards and mods, I had a similar train of thought though - could one take a standalone DVD player and connect from a point prior to the RGB-ification of the data directly to the parallel inputs on an 878 chip. Sort of a poor hacker's SDI, with a lot more solder.

 

[KBK]

yes. the output of the DVD drive is digital,and the questionis, what digital format? same format? if it is... hmmm.....


I wouln't myself, want to go the SDI route. I might find a way to fdo it for folks, but.. I would rather stay in the format that is out there. "ppor man's Parallel" if you will. Real poor. But fully functional. who needs the SDI hassle. And, the Zoltrix card is already laid out for it...With no SDI, you get the function..but at about $1k cheaper. just 8 wires, and the control fucntion..if any.Clocking might be aproblem, but if it syncs with the i2c clock,well,that's already there.


You see, that's where the missing connector between the chip and S-video input goes! it's a 8-bit paralell digital port!!


Since the card is pre-ported for the parallel function, it is just that much easier to do the SDI as well.

 

[luvmytivo]

Quote:

Originally posted by KBK yes. the output of the DVD drive is digital,and the questionis, what digital format? same format? if it is... hmmm.....

Pretty much pure MPEG2 data in ATA/IDE bus. Not CIR or anything that a cap card could understand. The more I think about it, the more I like the parallel->parallel solution - you bypass the 2 worst offenders in the PQ arena...

 

[siwilson]

I don't think what is coming off the DVD drive is actually MPEG. Isn't it a standard file read off the filesystem, just coming off in 1s and 0s. You would need to grab the MPEG file out of the filesystem data and then decode the CSS encrypted VOB files, which contain both the audio and video information. Then the decrypted video data would need to be formatted into frames based on the MPEG flags. I suppose that once you had the frame data you could then intercept it when it is written to the overlay and send it out another port of some kind.


Or am I completely missing the point

 

[JohnAd]

Ken


I think the nearest you would get for a one box solution would be to connect a Hollywood+ or a Creative DXR3 card to a bt8x8 card.


The DVD decoder cards almost certanly have CCIR656 floating around in it, probably feeding the video encoder chip.


All you need to do is ribbon that stream off the decoder card and into the GPIO of the input card.


This should work assuming the DVD player doesn't interfere with DScaler. Maybe a use for a dual processor. Here there should be no need to go via SDI which should keep the cost to two cards, time and a bit of ribbon cable.



John

 

[KBK]

So.. with possibly a drx3, and capture card, and a 1G Pentium...and a Radeon, what you get is a DVD/scaler/dscaler?


except for the conflict. on the bus..which would have to re resolved. I'm game.. but is it worth the hassle? Heck, with the drx3 handling ONLY the 656 conversion, and leaving the DVD drive alone.. this could work.


You could switch between device, depending on whether or not it was a bad DVD or not. So, you could use an outboard LD, VHS, Dish, etc.. or... Radeon and Cinemaster, or DVD drive and Radeon..and dscaler. your choice. And with the capture card.. the IRport is already there. just needs to be programmed and ported.