HDCP cracked

[valnar]

I wonder if this device along with a HDMI input device would be a benefit to Sage?
http://it.slashdot.org/story/11/11/2...opy-protection
http://aktuell.ruhr-uni-bochum.de/pm...m00386.html.en

[Fuzzy]

It would be a marginal improvement over component video capture for sources such as Cable/Sat STBs... expensive though, considering the $350 for this PLUS the Colossus capture card. For that cost, I'd much rather use the R-5000HD type mod that gets you the actual broadcast stream. The articles talk about it as a crack for bluray, but I think many hear would already MUCH prefer to use something like AnyDVDHD for that, and you end up with a MUCH better quality.

[valnar]

Originally Posted by Fuzzy

but I think many hear would already MUCH prefer to use something like AnyDVDHD for that, and you end up with a MUCH better quality.

Well obviously.... I own AnyDVD too. My only interest in the HDCP device is to decrypt HDTV. I'd guess that unencrypted HD would be too much for today's hard drives though. An offline recompression would be needed of course, but I don't know if any consumer PC could keep up with the initial raw 1080p video. Still, it's an interesting experiment.

[Fuzzy]

Originally Posted by valnar

Well obviously.... I own AnyDVD too. My only interest in the HDCP device is to decrypt HDTV. I'd guess that unencrypted HD would be too much for today's hard drives though. An offline recompression would be needed of course, but I don't know if any consumer PC could keep up with the initial raw 1080p video. Still, it's an interesting experiment.

The HD-PVR colossus DOES have an HDMI input on it, for non-protected capture, and encoding to H.264 (jsut as it does on the component inputs). This is how you would use it, but like I said, that would be MORE expensive than using an R-5000HD box, and would take up more hard drive space, and have lower quality.

[uberpixel]

I've already placed my pre-order on Amazon...

Okay, not really. But, for a reasonable price it sure would be nice to capture hdmi with my Colossus from my comcast (Motorola) cablebox...

I'm guessing that it will be a while before these hit the black market. Not sure what the legal implications would be for anyone making, selling, buying, or owning one of these devices?

Not sure that a clean digital picture of Project Runway is worth risking jail time...

Although my wife may disagree.

- uberpixel

[Fuzzy]

I'm not sure you'd be able to tell the difference between that and the component capture.

[valnar]

Originally Posted by Fuzzy

I'm not sure you'd be able to tell the difference between that and the component capture.

If you have a good TV, its easy to tell.

[wayner]

Originally Posted by valnar

If you have a good TV, its easy to tell.

Are you sure?

In this instance you will still be reencoding the video to H.264 with the Collosus card. It isn't like the difference between using the component outputs vs HDMI outputs of a BluRay player or STB. For TV signals it doesn't mean that you are getting a bit for bit representation of the original MPEG-2 or MPEG-4 stream as you are still depending on the quality of the H.264 compression algorithm/chip in the Collosus.

I believe the signal path for cable/sat would be like this:

• MPEG-2 (or MPEG-4) stream from cable/sat provider to STB
• STB decodes the stream and sends out an uncompressed, HDCP encrypted digital signal via the HDMI port
• This FPGA device would strip the HDCP encryption
• The uncompressed and unencrypted HDMI signal would go to the collosus which would then encode to H.264

Given the amount of compression that the H.264 encoder will entail will there really be any quality difference between using the component or HDMI input?

Theoretically you could capture the uncompressed, unencrypted signal but you are going to need a lot of hard drives! According to this webpage an uncompressed 1080i signal has a bitrate of 1.24Gbps. Therefore one hour of video would use 560GB. That's like two orders of magnitude more than the highest bitrate for the Collosus/HD-PVR.

[stanger89]

Yeah, remember HDMI isn't a video bitstream, it's raw video. The reason "digital" capture is better in the case of an R5000, or cablecard/QAM, is because it avoids the decode/video processing/re-encoding steps. Recording HDMI with HDCP stripped doesn't avoid any of that.

[uberpixel]

So how would this data path differ from my HDHR?

I get excellent quality video from the HDHR compared to washed out video from the Colossus. I though it was due to the component signal, but it sounds like you are saying it's due to the processing. Maybe I need to tweak my color settings or something?

- uberpixel

[panteragstk]

My colossus looks just as good as the original directv feed using hdmi (btw, if you don't want hdcp, directv doesn't use it). Directv is just as overly compressed as anyone else so it still looks OK at best. All TV with the exception of OTA looks OK to me. Some companies compress more than others for sure, but all of them have to.

[Fuzzy]

I have yet to even unpack either of my colossus cards, as my R-5000's with dish network are just SO much better. (Dish also does not use HDCP i believe). I'd rather have the head-end quality 5 Mbps H.264 1440x1080 from the R-5000, than that same thing, decompressed and scaled to 1920x1080 in the stb, then compressed by the colossus to consumer quality 12Mbps h.264 in the colossus.

[tmiranda]

Originally Posted by Fuzzy

I have yet to even unpack either of my colossus cards, ....

If I PM you my address you can ship them to me and I'll unpack them for you

[Fuzzy]

Originally Posted by tmiranda

If I PM you my address you can ship them to me and I'll unpack them for you

Nope... :-)

I got them when my R-5000's started acting up, just in case.. after some rearrangement of devices, I discovered the cause of the acting up was from overheating... with that fixed, I'm good on the R-5000's again.. not gonna give up my backup though.

[stanger89]

Originally Posted by uberpixel

So how would this data path differ from my HDHR?

The HDHomeRun, and all other digital tuners simply take the raw MPEG-2 (or H.264 in the case of an R5000 on Dish) transport stream and save it to disk. Nothing is changed, it's exactly as the broadcaster sent it.

Think of it like ripping a DVD or CD, you end up with the exact bits that were on the disc when you're done.

Recording from HDMI or component, the player/set top box receives that MPEG-2 transport stream, then demuxes and decodes the video into raw digital video, it's then sent over HDMI (or converted to analog) and send to the "display". If it's a TV, it just displays it, if it's an HDPVR or Collossus, it takes that raw digital video (converted from analog back to digital in the case of component), then runs it through an H.264 video encoder and creates a new transport stream with the audio/video.

Think of this like rather than ripping a DVD, recording from the S/Video output. You can get it back to the same format you had, but it's not the same bits.

Now, an HD PVR or Collosus quality is a lot better than S/Video, but the concept is correct.

I get excellent quality video from the HDHR compared to washed out video from the Colossus. I though it was due to the component signal, but it sounds like you are saying it's due to the processing. Maybe I need to tweak my color settings or something?

- uberpixel

That's where I'd start, or possibly even the output settings of your set top box.

The most likely artifact from an extra analog conversion is a softening of the image, a loss of detail. But Component is a very good transport, and from what I've seen the HD PVR is quite good. I'd say you're unlikely to notice any negative issues from the analog conversion.

If you're seeing a brightness change, or color changes, that would indicate a calibration issue.

Now as far as the extra decode/encode steps, it's actually exactly like ripping a DVD, or better yet a Blu-ray. You can convert it to another format, but there's some loss due to the process. If you convert to a high enough bitrate, then the loss is likely "insignificant".

Now, there are two wrenches to be thrown into the picture when using an HDPVR/Collosus vs the comparison to Blu-ray transcoding:
1) The source bitrate is often much, much lower. In the case of Dish, so much so that the HDPVR/Collosus recordings are much larger for the same basic quality.
2) Since the source bitrate is so much lower, especially with MPEG-2 sources (MPEG-4 handles insufficient bitrate more gracefully/subtlely in my experience) that the artifacts in the source drive up the required to bitrate even higher, potentially beyond what the HDPVR/Collosus are capable of meaning the recording exacerbates issues/artifacts.

Originally Posted by Fuzzy

I have yet to even unpack either of my colossus cards, as my R-5000's with dish network are just SO much better. (Dish also does not use HDCP i believe).

That's not been my experience, between my R5000 and my HD PVR, I can't tell which recorded it. I let Sage manage which is used and I can't tell just from starting a show which encoder recorded it.

Well that's not entirely true, if there's comskip files for the recording, the R5000 didn't record it, SA seems to bomb on R5000 recordings. But even when I've recorded the same thing on both R5000 and HD PVR I've not seen a difference in quality between the two. At least not with Dish's streams played back through my extenders on my projector.

I'd rather have the head-end quality 5 Mbps H.264 1440x1080 from the R-5000, than that same thing, decompressed and scaled to 1920x1080 in the stb, then compressed by the colossus to consumer quality 12Mbps h.264 in the colossus.

Now the size is definitely a knock on the HD PVR vs the R5000, but the HDPVR recordings seem a lot more compatible than the R5000 ones.

[BobPhoenix]

Originally Posted by wayner

Given the amount of compression that the H.264 encoder will entail will there really be any quality difference between using the component or HDMI input?

I have a Colossus upstairs recording HDMI and 2 HD-PVRs downstairs recording Component from DirectTV. I see a difference. Very slight but there is a difference. If I didn't know it was a Colossus I might not be able to tell without examining the picture allot more than I need or want to.

[Fuzzy]

That's not been my experience, between my R5000 and my HD PVR, I can't tell which recorded it. I let Sage manage which is used and I can't tell just from starting a show which encoder recorded it.

Well that's not entirely true, if there's comskip files for the recording, the R5000 didn't record it, SA seems to bomb on R5000 recordings. But even when I've recorded the same thing on both R5000 and HD PVR I've not seen a difference in quality between the two. At least not with Dish's streams played back through my extenders on my projector.



Now the size is definitely a knock on the HD PVR vs the R5000, but the HDPVR recordings seem a lot more compatible than the R5000 ones.

Comskip has no problem handling my dish R5000 files, and I used to use Sa, and it always seemed to work as well. I believe you that the quality is similar, I wouldn't have bought it if it wasnt good enough. The size thing is pretty significant to me though.

[uberpixel]

Originally Posted by BobPhoenix

I have a Colossus upstairs recording HDMI and 2 HD-PVRs downstairs recording Component from DirectTV. I see a difference. Very slight but there is a difference. If I didn't know it was a Colossus I might not be able to tell without examining the picture allot more than I need or want to.

I just replaced an older Vizio 720P LCD with a new Samsung 1080P LED set, so that is probably helping to accentuate the difference, but I can tell that the Colossus is both softer and less saturated.

I've replaced my component cables as well, since I had some kind of scrolling noise that was likely due to a ground fault or something and that seems to have lessened the discrepency between my tuners.

Anyway, thanks for your feedback with respect to the Colossus. I think I derailed the thread somewhat, but I suppose this is all still relevant.

- uberpixel