Porting SageTV to run on D-Link DSM-520

[blade]

We've all been wanting a HD extender, but I haven't seen anyone say how the quality compares to the Nvidia cards with PureVideo. Not necessarily with HD content, but SD scaled to 1080i. I assume this device can do that? Will SD shows captured and seen as video be correctly deinterlaced with devices such as these? Currently the only way I can get proper deinterlacing is nvidia purevideo decoder with smart deinterlacing.

[GbrNole]

blade - that's a damn good point actually.

my mvp provides a better picture than my 6600gt pure video setup does strictly for SD recordings - it's not massively better but it is better but like you say i wonder how it would look if it was a scaled image?

i think more than anything i like the mvp because i don't have to worry about drivers, OS etc. at the client end. just the server to mess with.

[Scriber]

Originally Posted by GbrNole

my mvp provides a better picture than my 6600gt pure video setup does strictly for SD recordings - it's not massively better but it is better

If I could take this a step further and not too OT..... I'm curious, how does the MVP's PQ for SD recordings on a HDTV via Svideo compare with the 6600's PQ on an HDTV via VGA? I presumed that the 6600's PQ had to be superior due to resolution and superior connection but perhaps not given the HW decoder.

[GbrNole]

Originally Posted by Scriber

If I could take this a step further and not too OT..... I'm curious, how does the MVP's PQ for SD recordings on a HDTV via Svideo compare with the 6600's PQ on an HDTV via VGA? I presumed that the 6600's PQ had to be superior due to resolution and superior connection but perhaps not given the HW decoder.

i pretty sure i answered that?

the MVP looks better than the 6600gt when the 6600gt is spitting out 1366x768p to my plasma over vga and the mvp is connected via svideo.

now for good quality SD images it gets to become a toss up (ie. dvd) but for recorded tv the MVP is better. a bit better actually.

[JUC]

I've got my MVP driving a 42" plasma and the quality is really good for me...

[Scriber]

Originally Posted by GbrNole

i pretty sure i answered that?

the MVP looks better than the 6600gt when the 6600gt is spitting out 1366x768p to my plasma over vga and the mvp is connected via svideo.

now for good quality SD images it gets to become a toss up (ie. dvd) but for recorded tv the MVP is better. a bit better actually.

Hmmmmm...I guess i'm a bit slow then and I wasn't clear you meant HDTV.

Even so, if the MVP is better than a 6600 displaying SD on an HDTV (per your experience) then other devices (assuming similar decoders) should too - right? So Blade's concern is probably/hopefully a non-issue then....? Or am I assuming too much when i presume other mfrs will use a decoder of the MVP quality or better?

[blade]

I've always heard a good video card such as the 6600 did a better job than the MVP when using a HD display. Maybe some more people will chime in, because that's the main reason I didn't go with the MVP when the extender client first came out. Now I feel I need to hold off for a possible HD capable extender instead of throwing money into an old MVP.

[AngelofDeth]

Originally Posted by Scriber

Hmmmmm...I guess i'm a bit slow then and I wasn't clear you meant HDTV.

Even so, if the MVP is better than a 6600 displaying SD on an HDTV (per your experience) then other devices (assuming similar decoders) should too - right? So Blade's concern is probably/hopefully a non-issue then....? Or am I assuming too much when i presume other mfrs will use a decoder of the MVP quality or better?

An HD Extender would presumably work just as well as an MVP on SD content. It seems to me, that the MVP isn't so good because it has a hardware MPEG decoder, but the fact that it doesn't have to deinterlace/scale/re-interlace. DVD film material from a PC looks pretty decent even on my old crappy radeon 7000 with composite output.

[stanger89]

Originally Posted by blade

I've always heard a good video card such as the 6600 did a better job than the MVP when using a HD display. Maybe some more people will chime in, because that's the main reason I didn't go with the MVP when the extender client first came out.

The MVP is great if you're stuck in SD land recording to output.

But you can have my 6800 or 6600GT when you pry them from my cold dead hands. My 6800 decimated (OK, way overstating it), everything else I've tried for an HD display (720p FP):
Pioneer Elite 480p DVD Changer
MyHD MDP-130

Better deinterlacing, scaling, etc, just an overall better picture. My 6800 blows away the Elite and the MyHD on DVDs, and it bests the MyHD on HD as well. From when I tested my stuff with the HQV Benchmark DVD, the 6800 was the best thing I had, with the Faroudja chip in my PJ being second IIRC doing better on some things, and worse on others.

As of this second, I'd rate hardware accellerated playback ("PureVideo") on par with basic Faroudja processing overall (both have strengths). For my main display, I won't get too excited unless whatever solution has something HQV or Genum VXP level for processing, or that it provides "native" digital output for input into an HQV or Genum VXP processor.

[reboot_this]

Any word on this topic? updates? I guess I'm kinda getting ansy from all the "checking for v 5.1"

[JUC]

Nothing new unfortunately. I've been in contact with them on and off for the past two weeks. I usually try to get an update on fridays...might as well request one today. I think it has turned into the same old thing as before--waiting for sigma/redsonic....which really stinks. I'll keep putting some pressure on them and let you all know how it is going.
Juc

[stefano540]

They may have been busy with other legal issues: http://gpl-violations.org/news/20060...frankfurt.html
Looks like they don't understand the concept of GPL: "In the court proceedings, D-Link claimed that the GPL is not legally binding".

Hopefully they are now "sensitized" and will help out
Stefano

[dagar]

Wow, stupid is as stupid does. Nice find!

[Mahoney]

Sounds like it might be worth contacting GPL Violations, they would probably be very interested to hear about the lack of response from D-Link and it might speed things up a bit.

[Narflex]

Sarbanes-Oxley makes this even more interesting for publicly held US companies:

http://www.wasabisystems.com/gpl/sox.html

and of course the other side:

http://www.softwarefreedom.org/publi...nes-Oxley.html

[JUC]

Hot topic lately. As far as the German GPL issue---that is good news I think because it may put Dlink on the defensive in terms of bringing additional products of theirs into compliance.

AFA the article--I don't see it that the court only awarded gpl-violations.org 300 euros--but that the court found that their requested damages were excessive by the arbitrary amount of 300 euros. So the court awarded damages of 300 euros less than what was requested. Big difference in terms of payout if that is the case.

Re SOX--i'm going to stay out of that one as I have zero experience in the topic. However, at face value--the text quoted above seems to read clearly--failure to comply with the terms of the GPL means a potential violation of the Act. But--there is much more to the Act than what is posted above and I would bet that it is not as cut and dry as it seems. That's all i'll comment on for now.

I was thinking that maybe we should contact gpl-violations.org so that they can add this current and ongoing violation to their list for Dlink? Let me know and I can make the contact.

Can someone also answer me this--what is the harm for sigma/redsonic to release their source code? Why are companies so adverse to doing so if it doens't include any of their IP? That is one think I haven't really understood so far. I can see for small companies the documentation might be burdensome but for larger ones that have X amount of employees working on these things it doesn't seem like it would be that big a deal...
Thanks,
JUC


edit: hey-i just realized (should have figured it out sooner) that gpl-violations.org is not based in the states....is this product distributed in europe by any chance?

[PGPfan]

Originally Posted by JUC

Can someone also answer me this--what is the harm for sigma/redsonic to release their source code? Why are companies so adverse to doing so if it doens't include any of their IP? That is one think I haven't really understood so far. I can see for small companies the documentation might be burdensome but for larger ones that have X amount of employees working on these things it doesn't seem like it would be that big a deal...
Thanks,
JUC

My guess would be that they are afraid that they would be resposible (somehow?) for technical support of 'modded' machines that are rendered unuseable due to 'hacking'. Not a good arguement, but possible if viewed from a paranoid manufacturers perspective.

-PGPfan

[KrisAcker]

Originally Posted by JUC

hey-i just realized (should have figured it out sooner) that gpl-violations.org is not based in the states....is this product distributed in europe by any chance?

Yes, it's distributed in europe. A quick google for the DSM-520 limited to "Only pages from Belgium" shows a lot of Belgian shops selling it.

The reason in my opinion why they don't release the source code, is that it's "cheaper" not to. They will have estimated their liabilty, and booked that amount as a cost (possibly even in a previous financial year). Or they might be writing off the cost of over the lifetime of the product with little impact on each year's results. Unless the cost of not releasing the source code is going to get higher then the estimated amount, they won't worry about it because it no longer impacts on their P&L (thats profit & loss statement for the non-accountants among you). Should they need to revise the estimate because of greater damages awarded in similar cases, they might consider that releasing the code would be cheaper than booking an additional cost in their P&L.

Even better would be that they couldn't put a decent estimate on the amount of the cost anymore. Shareholders really don't like knowing that there will be costs in the future, but that nobody can tell them how high those costs will be. This kind of liability doesn't do much from a company's credit rating either. Usually this get's the management looking for a solution that will freeze the amount of the cost immediately.

[sleonard]

Originally Posted by JUC

Can someone also answer me this--what is the harm for sigma/redsonic to release their source code? Why are companies so adverse to doing so if it doens't include any of their IP?

I'm sure they consider any modifications that they write into the GPL code as "their" IP and thus their reluctance to release it, esp. when that code concerns hardware that they want to keep proprietary.

Scott

[blade]

I don't think SOX will have the effect on gpl violations that some people think it will. The purpose of SOX is to hold top executives more accountable for creating and maintaining strong internal controls. Ultimately the goal is to create more accurate financial statements and protect investors. Accounting isn't about perfection, it is about providing information in a format that is useful to reasonable users of the financial statements. The more potential something like gpl violations have to affect the finanacial statements the more attention it will get from the accounting world.

For anyone that is interested HealthSouth was the first case tried under SOX and we all know how that worked out.