Equivalent processor [Archive]

mlobitz

12-11-06, 11:41 PM

O.K., I'm finally putting this thing through the paces now that it hasn't frozen up on me since the complete reformatting of the hard drive. Now some questions...jump in here any time Earl...

1)What equivalent processor is this box, the HR20, using? Like a P4 and what ghz? Just a ballpark is fine. Reason I ask is that it chokes on two MP4s being recorded while doing other menu/guide manuevers. I did notice FOX 40, yes I was wrong, we now have FOX 40 in Sacramento on the MP4. Watched House tonight, not bad, not bad, looked pretty good. No OTA to compare, but please, it looked good. While I was recording that and a show on local NBC, I don't know what I just recorded it for fun, I went to the guide to see what time the Laker game was going to start tomorrow and then the fun began...audio drops and jagged picture in the little window.

I've played around with encoding mp2 video into DIVX and other MP4 codecs on my computer for years and the one thing I noticed is that you can't sneeze near the thing without it choking on the mp4 data. Touch the keyboard and the audio goes out of sync or if you are encoding the computer freeze. I have a P4 running at 2.4 ghz with a gig of memory. Plenty to do the encoding if you leave it alone. Are there enough horses in the HR20 to handle all the mp4 data? Dumb question, because why would D* release a box that would be too slow to hand that data.

2) Is D* sending some kind of info down the MP4 stream not related to the current programming and are they sending it only on say the FOX mp4 stream. I don't know...some background guide info or weather info. In a pulse that lasts say...I don't know 7-8 seconds? Just grasping at straws trying to understand what in the world happens on FOX...

D.L.B. and you got me...:)

jonaswan2

12-12-06, 01:41 AM

You can't really compare processors for CE devices with computer processors, but the MPEG4 decoding and general processing are done on seperate chips.

mlobitz

12-12-06, 07:25 AM

You can't really compare processors for CE devices with computer processors, but the MPEG4 decoding and general processing are done on seperate chips.

You mean like a main processor (P4) and a video processor...(like an ATI card, etc...) I think you can make a "ballpark" comparison, it seems like it's just that no one knows what it is...yet...

Any answers to the other questions?

laxcoach

12-12-06, 07:30 AM

walters

12-12-06, 07:35 AM

It would be a meaningless bit of trivia. Consider that the Series 1 standalone TiVo had a 54 MHz CPU (no, that's not a typo) and never had any problem encoding one stream and decoding another (because the heavy lifting is handled by dedicated hardware). The Series 1 DTiVo upped that to around 70-80 MHz (and doubled the memory) to be able to handle dual tuners (even though they were already compressed).

For whatever it's worth, though, I believe they used the same processor as that found in the Series 3 TiVo: Broadcom BCM7038:
http://www.broadcom.com/products/Cable/Digital-TV-Solutions/BCM7038

A 300 MHz 64-bit MIPS (along with a lot of other stuff). That's 0.3 GHz for you kids out there :D

mlobitz

12-12-06, 07:37 AM

Your question is a non-issue:
1) The HR20 DOES NOT encode data. It is encoded at the source. The HR20 records the bitstream directly from the satellite dish.
2) The DECODING is done by chips that have 1 function: Decode MPEG2/4.

The general purpose chips have no impact on recording satellite bitstreams or decoding the bitstreams from disk.

That's so funny that you state it that way as the HR20 has soooooooooo many issues that I don't know even where to begin.

Do the chips in the box have to chew a little harder on the MP4 stream than they do on say the MP2 stream or is it all just raw data to it? Why do the MP4 channels choke and die, especially when you try to record two of them, more often than the MP2 channels? Maybe it's not chip dependent, but something is grinding inside that box...

iacas

12-12-06, 07:43 AM

You can't really compare processors for CE devices with computer processors, but the MPEG4 decoding and general processing are done on seperate chips.

Ditto.

mlobitz

12-12-06, 07:55 AM

Ditto.

Ditto = I don't know

iacas

12-12-06, 08:03 AM

Ditto = I don't know

Uhm, no, it doesn't. You could look up the definition (hint: "as already stated; likewise.") and retract your statement or simply stop being a jerk.

That's so funny that you state it that way as the HR20 has soooooooooo many issues that I don't know even where to begin.

You've had a lot of issues. Again, many other people have not.

Do the chips in the box have to chew a little harder on the MP4 stream than they do on say the MP2 stream or is it all just raw data to it?

In fact, when writing to the disk, the MPEG-4 is easier to write because it's easier to compress. So, all other things being equal (i.e. the picture, etc.) MPEG-4 is "less intensive" to write to disk than MPEG-2. There are simply less 1s and 0s. Less data to write. Lower bitrates... that's what makes MPEG-4 "better" than MPEG-2.

Why do the MP4 channels choke and die, especially when you try to record two of them, more often than the MP2 channels? Maybe it's not chip dependent, but something is grinding inside that box...

Your experiences are likely not representative of the whole, the majority, or of even a large percentage of the minority.

MPEG-4 is often a little more difficult to decode, but not always and not by much. If you want more, use a search engine to answer your questions.

LameLefty

12-12-06, 08:15 AM

Do the chips in the box have to chew a little harder on the MP4 stream than they do on say the MP2 stream or is it all just raw data to it? Why do the MP4 channels choke and die, especially when you try to record two of them, more often than the MP2 channels? Maybe it's not chip dependent, but something is grinding inside that box...

My MPEG4 streams have never given my box any problems systematically - our CBS local HD is glitchier than the others but it's never crashed my box. In fact, the three GUI lockups we've had over the last 7-1/2 weeks have all come when selecting MPEG2 SD programs to play from the Playlist or telling it to delete an MPEG2 SD program at the end of playing. No HD programs, whether MPEG2 or 4, have been central to any of our real issues with the box.

mlobitz

12-12-06, 08:15 AM

Uhm, no, it doesn't. You could look up the definition (hint: "as already stated; likewise.") and retract your statement or simply stop being a jerk.

You've had a lot of issues. Again, many other people have not.

In fact, when writing to the disk, the MPEG-4 is easier to write because it's easier to compress. So, all other things being equal (i.e. the picture, etc.) MPEG-4 is "less intensive" to write to disk than MPEG-2. There are simply less 1s and 0s. Less data to write. Lower bitrates... that's what makes MPEG-4 "better" than MPEG-2.

Your experiences are likely not representative of the whole, the majority, or of even a large percentage of the minority.

MPEG-4 is often a little more difficult to decode, but not always and not by much. If you want more, use a search engine to answer your questions.

Please:lol: stop it...:lol: can't stop:lol: laughing:lol:

It amazes me. People give ME short curt little answers like, "Your question is a none issue..." instead of writing what you just wrote above which actually ANSWERS PART OF MY ORIGINAL QUESTION!!!

Thanks for calling me a jerk and I guess for answering my question?

O.K. I'm back on the floor now:lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol:

You people take yourselves way too seriously. Even when Ironcrap Com. missed four of my appts., yes I was angry, but not a jerk...

Here it comes again...:lol: :lol: :lol: :lol: :lol: :lol:

Doug Brott

12-12-06, 08:18 AM

For whatever it's worth, though, I believe they used the same processor as that found in the Series 3 TiVo: Broadcom BCM7038:
http://www.broadcom.com/products/Cable/Digital-TV-Solutions/BCM7038

DirecTV uses the BCM7411 chip to decode MPEG2/MPEG4. Not sure of the processor because I didn't open the box. The processor should be under the heat sink at the top of this photo. This picture are from Earl.

http://www.dbstalk.com/hr20/images/Small/20-underharddrive.jpg

I checked out the Product Brief (http://www.broadcom.com/collateral/pb/7411-PB05-R.pdf) from Broadcom and on page 2, the following statement is made:

The BCM7411 supports the following decoding operations:
* 1 HD (MPEG-2, H.264 or VC-1) in real-time
* 2 SD (MPEG-2, H.264 or VC-1) in real-time

H.264 = MPEG4

Doug Brott

12-12-06, 08:23 AM

In fact, when writing to the disk, the MPEG-4 is easier to write because it's easier to compress. So, all other things being equal (i.e. the picture, etc.) MPEG-4 is "less intensive" to write to disk than MPEG-2. There are simply less 1s and 0s. Less data to write. Lower bitrates... that's what makes MPEG-4 "better" than MPEG-2.

The steam coming from DirecTV is already compressed. I think what you meant to say is that MPEG4 is more compressed than MPEG2, thus taking less space on the disk. The HR20 never encodes the data, it only decodes it. MPEG4 is actually significantly harder to compress to than MPEG2. It's only been in the last few years that people have been able to encode data into MPEG4 in real time making the currently technology possible.

MPEG-4 is often a little more difficult to decode, but not always and not by much. If you want more, use a search engine to answer your questions.

This sounds about right.

O2BRich

12-12-06, 08:42 AM

Doug Brott

12-12-06, 09:16 AM

Doesn't the HR20 only decode the MPEG compression for viewing?

Doesn't it actually store the raw feed to HD?

If that is true, then wouldn't it take very little CPU and no MPEG chip resources to record.

However to watch would be a different issue.

In fact in another discussion about PIP the same picture was shown and stated since it only had one decoder PIP could not work.

Correct, the "record" function basically reads the data stream from DirecTV or OTA and puts the data directly onto the hard disk drive. DirecTV already knows that this format will be either MPEG2 or MPEG4 because either they encode it themselves or they are receiving a known format over-the-air. There is ZERO encoding on the computer, so there is no need for an encoder chip or a larger processor to handle the encoding.

wheelswagz

12-12-06, 09:47 AM

DirecTV uses the BCM7411 chip to decode MPEG2/MPEG4. Not sure of the processor because I didn't open the box. The processor should be under the heat sink at the top of this photo.

I checked out the Product Brief (http://www.broadcom.com/collateral/pb/7411-PB05-R.pdf) from Broadcom and on page 2, the following statement is made:

This is one of the most direct and informative answers I have read on DBSTalk! Thanks brott!

mlobitz

12-12-06, 10:30 AM

DirecTV uses the BCM7411 chip to decode MPEG2/MPEG4. Not sure of the processor because I didn't open the box. The processor should be under the heat sink at the top of this photo. This picture are from Earl.

http://www.dbstalk.com/hr20/images/Small/20-underharddrive.jpg

I checked out the Product Brief (http://www.broadcom.com/collateral/pb/7411-PB05-R.pdf) from Broadcom and on page 2, the following statement is made:

H.264 = MPEG4

Dude...you are awesome. See, the others who didn't really have the info to answer the question just c*apped on me and called me a jerk. You...you waded through and came up with the answer. I thank you for that. I've had some bad experiences with this box and I will be getting a new one, but for the time being I'm just trying to make sense of it all including why D* even released it in the first place. Not trying to offend, but sometimes people who are direct and to the point come off like that.

Thanks again!

Just as an aside, a lot of these "high powered" gaming systems (XBOX, PS2) use 300 mhz processors.

Doug Brott

12-12-06, 10:38 AM

Dude...you are awesome. See, the others who didn't really have the info to answer the question just c*apped on me and called me a jerk. You...you waded through and came up with the answer. I thank you for that. I've had some bad experiences with this box and I will be getting a new one, but for the time being I'm just trying to make sense of it all including why D* even released it in the first place. Not trying to offend, but sometimes people who are direct and to the point come off like that.

Thanks again!

Just as an aside, a lot of these "high powered" gaming systems (XBOX, PS2) use 300 mhz processors.

No worries. We all just want more information.

walters

12-12-06, 10:41 AM

It's like my post doesn't even exist :(

:D

Doug Brott

12-12-06, 11:23 AM

It's like my post doesn't even exist :(

:D

Yeah, I'm trying to track down whether or not the processor is the BCM7038 or not. I found a post over at TCF that indicated the BCM7038 was in use, but I'm still a doubter - the post also stated that there was only ONE ATSC tuner in the HR20, which is clearly not true from the pictures of the hardware. The BCM7038 has some of the same decoding capabilities of the BCM7411 (it may be less efficient, though). It also has the MIPS processor. It is possible that DirecTV has disabled the decoding routines in the BCM7038 and is doing it externally with the BCM7411. I suspect that a different processor is in use, but I do not know that one way or the other.

lguvenoz

12-12-06, 11:30 AM

Just as an aside, a lot of these "high powered" gaming systems (XBOX, PS2) use 300 mhz processors.

As a game system fan, not a fair statement... The original XBox was a 733MHz Intel chip, and the new platforms are 3.2GHz cores (Xbox360 with 3 cores running 2 parallel instructions per core and PS3 utilizing the Cell processor which runs at comparable speeds) supporting in excess of 1 teraflop between the CPU and GPU chips.

lguvenoz

12-12-06, 11:33 AM

Correct, the "record" function basically reads the data stream from DirecTV or OTA and puts the data directly onto the hard disk drive. DirecTV already knows that this format will be either MPEG2 or MPEG4 because either they encode it themselves or they are receiving a known format over-the-air. There is ZERO encoding on the computer, so there is no need for an encoder chip or a larger processor to handle the encoding.

Does anyone know how D* is handling the encryption of the data as it is streamed to the disk?? It is my understanding that the raw stream is actually encrypted and decrypted when writing and reading from the drive. Just curious as to if/how this is being done.

mlobitz

12-12-06, 11:46 AM

As a game system fan, not a fair statement... The original XBox was a 733MHz Intel chip, and the new platforms are 3.2GHz cores (Xbox360 with 3 cores running 2 parallel instructions per core and PS3 utilizing the Cell processor which runs at comparable speeds) supporting in excess of 1 teraflop between the CPU and GPU chips.

I stand corrected, I'm pretty sure either the PS2 or XBOX was only 300 mhz, but if you say 733, it's all good. Just saying that these boxes are not as high powered as they seem.

Also, Walters, you rock too man. Thanks for the info.

Also, could the encryption slow down the system? Does it vary from stream to stream, different types of encryption? I'm just trying to figure out where this thing is hanging up.

lguvenoz

12-12-06, 11:56 AM

I stand corrected, I'm pretty sure either the PS2 or XBOX was only 300 mhz, but if you say 733, it's all good. Just saying that these boxes are not as high powered as they seem.

Also, Walters, you rock too man. Thanks for the info.

Also, could the encryption slow down the system? Does it vary from stream to stream, different types of encryption? I'm just trying to figure out where this thing is hanging up.

Just giving a bit of ribbing on the game machines :) The new ones are on par with super computers from 5-6 years ago in terms of raw processing though. It's pretty scary.

The encryption is a great question. I would hope they did something akin to the latest PCs. They are equipped with what is called a TSM chip that stores the encryption key and is responsible for encryption and decryption of data streams. I believe this is something that ViiV mandates, is part of a lot of the new AMD systems, and is something the Windows Vista definitely embraces for securing information in transit (both network and within the system data bus) and at rest (on disk).

Canis Lupus

12-12-06, 12:48 PM

iacas

12-12-06, 01:00 PM

The steam coming from DirecTV is already compressed. I think what you meant to say is that MPEG4 is more compressed than MPEG2, thus taking less space on the disk.

Yes, "easier to compress" is poorly phrased. I meant to say - as the rest of that paragraph indicates - that MPEG 4 is MORE compressed. Not "easier." Being more compressed makes it "easier" to write because the same "stuff" (say, a one-hour TV show) is less data, thus meaning there's less to write, thus making it a little "easier" for the recorder to record the show.

iacas

12-12-06, 01:09 PM

It amazes me. People give ME short curt little answers like, "Your question is a none issue..." instead of writing what you just wrote above which actually ANSWERS PART OF MY ORIGINAL QUESTION!!!

Thanks for calling me a jerk and I guess for answering my question?

Your original question was irrelevant. Processor speed in a CE device cannot be compared to processor speed in a fully functional computer. It's not a valid question.

I quoted the original person who said "not a valid question" and said "ditto." You then told me I "didn't know." I knew, and was telling you the question was irrelevant.

I stand corrected, I'm pretty sure either the PS2 or XBOX was only 300 mhz, but if you say 733, it's all good. Just saying that these boxes are not as high powered as they seem.

That ignores the graphics processors, speed of the RAM, texture caches, dedicated physics processors, vector units, and all sorts of other things game systems makers _might_ add to their system(s) to make them function properly. Again, it's irrelevant to compare the processor speeds of different devices performing wholly different tasks.

You can't even compare processors in computers. MIPS processors, IIRC, are RISC (as are PowerPC processors). They are - generally - capable of doing more per cycle than a CISC processor. It's an irrelevant line of questioning, and none of the answers will help you figure out why your box is more fluky than just about everyone else's.

Also, could the encryption slow down the system? Does it vary from stream to stream, different types of encryption? I'm just trying to figure out where this thing is hanging up.

Your box does not do encryption, nor does it do any compression. Your box decompresses video - "codec" stands for "compression-decompression." Your box only handles the "decompression" side of things. MPEG-2 and MPEG-4 data are all the box gets. That's all it likely knows how to decompress.

You have a bad box, and getting side-tracked into odd questions isn't the way to go.

Just giving a bit of ribbing on the game machines :) The new ones are on par with super computers from 5-6 years ago in terms of raw processing though. It's pretty scary.

Depends on your definition of "supercomputer." A desktop Mac from several years ago was technically a "supercomputer." Remember the Apple ad with the tanks?

http://youtube.com/watch?v=SRlstV2tjwc

"a billion calculations per second" is a pretty old definition. A more thorough discussion can of course be found here: http://en.wikipedia.org/wiki/Supercomputer .

mlobitz

12-12-06, 01:56 PM

I don't know if you've stepped through this or not, but have you checked your signal strengths for the SAT where you get your locals? Seems many of the hangups and probs you're describing for your MPEG-4s may be an issue of signal strength, a bad multiswitch, or a bad LNB for the 99 or 103. Apologies in advance if you've covered these possibilities already.

I could list all the transponder strengths, but they are 80-100 across the board. The local spot beam 101 #4 is 100%. The guy who finally did my install from Ironc*ap did a great job. No it's not that...I checked it first thing. When the install guy comes, I do half or more of the work setting it up. Everything is good, my HR10-250 hasn't missed a single recording EVER. That's why this thing kills me. It doesn't do even its most basic functions well.

Hey iacas...did you see that South Park where they play World of Warcraft against that one all powerful guy in the game? Dude...that's you dude...If you haven't seen it, youtube it, it's a great episode. Of course, the SP boys totally beat him in the end. Awesome...

On a serious note, I have been called a jerk in this forum, but man you are a technojerk. What an a**. I asked a simple question regardless of whether my box is good or fluky(sp?) what is that...anyway, not the point.

richlife

12-12-06, 02:38 PM

Moderator -- can we derail this thread? There's way too much name calling and not enough positive.

mlobitz, you don't seem to want anything except "direct" answers that noone can give you. I said it before, your box exhibits unique as well as non-unique problems that seem more likely (to me) indicative of manufacturing defects (maybe the assembler had a cold and you get the virus :<) than a set of software defects that only you have found -- regardless of how many code paths you pounded through.

Meklos

12-12-06, 02:46 PM

Your box does not do encryption, nor does it do any compression.

Actually it does encryption, I'm pretty sure. I'm sure they based the current methodology off of the Tivo "style" of doing things, which involved having a crypto chip onboard each unit with a unique key, so that recordings made on that unit can only be played back on that unit.

If they didn't do this, the content providers would be worried that the video data would be insecure.

Doug Brott

12-12-06, 02:57 PM

Moderator -- can we derail this thread? There's way too much name calling and not enough positive.

agreed.

mlobitz

12-12-06, 03:00 PM

I agree as well. Uh, hey wait!

jonaswan2

12-12-06, 03:02 PM

Meklos

12-12-06, 03:04 PM

I don't think there is any encryption done in the box, but I could be very wrong. I do know that they handle encryption the VideoGuard way and not the Tivo way.

If this is the case, then there's trouble ahead. If the data on the drive isn't encrypted in some way, then that show is as good as posted to some site.

Surely they're encrypting the data on the drive?

LameLefty

12-12-06, 03:07 PM

Surely they're encrypting the data on the drive?

They have to be - otherwise everyone who's hooked up an ESATA drive would be crowing about it. The data streams are encrypted and reverse-engineering it is now illegal under the DMCA 'cause everyone knows that consumers are just pirates and have no legitimate interest in the stuff they record except to copy it a billion times and release it on the internet, costing those poor huge corporations billions of dollars in additional revenue. :rolleyes:

mlobitz

12-12-06, 03:09 PM

O.K. I spoke to a "real" people who is pretty tech savvy and we both were :lol: . All I asked was could someone give me a comparative speed on the processor and how it managed the data and whether that could be part of the problem with audio dropouts, et. al. The responses have been hilarious. Anything from helpful to my geek is bigger than your geek. Whatever. I don't want direct answers! I said BALLPARK answers that's all. Ballpark means close to the answer, if any will do. Some people have felt attacked by my directness, not direct questions, but so many random people have attacked me that I just have to :lol:

Seriously, you should check out the World of Warcraft episode. Hilarious. and you can see iasasas or whatever that name is...

laxcoach

12-12-06, 03:11 PM

That's so funny that you state it that way as the HR20 has soooooooooo many issues that I don't know even where to begin.

Do the chips in the box have to chew a little harder on the MP4 stream than they do on say the MP2 stream or is it all just raw data to it? Why do the MP4 channels choke and die, especially when you try to record two of them, more often than the MP2 channels? Maybe it's not chip dependent, but something is grinding inside that box...

You completely misunderstood my statement. I meant that your question assumes things that are untrue. You asked about the equivalent computing power in the context of encoding/decoding mpeg4. But that bitstream doesn't get worked on by the "cpu". It is never encoded on the box, and when it is decoded it is by hardware.

Reason I ask is that it chokes on two MP4s being recorded while doing other menu/guide manuevers.

There is absolutely no analogy between your desktop and the architecture of a machine like this.

My answer had nothing to do with overall architecture issues. It was purely an answer to your assumptions. Your tone may be meant to be funny, but it reads as obnoxious. I'll soooooo give you the benefit of doubt though.

jonaswan2

12-12-06, 03:11 PM

If this is the case, then there's trouble ahead. If the data on the drive isn't encrypted in some way, then that show is as good as posted to some site.

Surely they're encrypting the data on the drive?

When you get the raw stream, it's already encrypted with VideoGuard and can only be viewed with a compatible smart card from NDS/DirecTV, then the DVR just saves that data (I actually think that's how NDS's VideoGuard DVR works).

Meklos

12-12-06, 03:14 PM

When you get the raw stream, it's already encrypted with VideoGuard and can only be viewed with a compatible smart card from NDS/DirecTV, then the DVR just saves that data (I actually think that's how NDS's VideoGuard DVR works).

AH of course, brain fart. Had to go read up on VG again...

Meklos

12-12-06, 03:21 PM

There is absolutely no analogy between your desktop and the architecture of a machine like this.

It's like asking how wide the tape on an analog VCR would have to be to record and playback a digital signal at 1080i.

It takes a very modern PC and relatively high end video card to playback full 1080i HD. I'm gonna venture a guess and say that the RCA DTC-100 doesn't have a P4 2.8GHz nor a 256M ATI 1800-class video processor in it.

Not just apples and oranges. Apples and sewing machines.

[edit] The reason people are responding this way is that the 'speed' of the machine can have no bearing on the set of problems that mlobitz is describing / has described. We can tell you how fast the processor is (if we know), then he can turn around and say it's too slow that must be the problem it's a piece of junk blah blah. It's a whole different set of capabilities, and processor speed isn't going to affect actual video playback. It might affect menu draw speed, etc... but not actual video - since it's not traversing the CPU (unlike a modern PC).

Clint Lamor

12-12-06, 03:26 PM

Ok any and all name calling will cease, if it doesn't then we will deal with the users doing so. If someone starts calling you names DO NOT respond by calling them names. REPORT the post and we will deal with those users. You have all seen numerous posts on this and it's no longer acceptable, there is a sticky at the top addressing this behavior I have addressed it in a few posts and it's getting quite old. It's going to stop or there are going to be more vacations from here handed out.

Clint Lamor

12-12-06, 03:31 PM

Oh and by the way 300 MHz in the non x86 world is FAR different the 300 MHz in the x86 world. Just look at most of the older Sun, SGI type of machines they ran a low speed RISC processor at much slower speeds then the Intel crowd and would in most instances smoke them.

mlobitz

12-12-06, 03:55 PM

I think he was just asking a questions trying to get some basic information on some of his problems with the HR20-700. Instead many have bashed him and told him how crazy his question is/was. That is not very nice, simply explaining that it cannot or will not work would be fine. I see that many of you are very technology savvy. Maybe Directv could use you to help expedite the updates on the HR20-700.

Uh, thanks, I appreciate at least one person seeing this with anything other than geek eyes. Not name calling just saying...anyway, seems to be many conflicting ideas on how this thing works. No need to continue. Since I'm the one who started this thread and I basically know now that no one really knows, then cool. No biggie, I don't really care that much. Just a question. Maybe I'll just keep my post to the following...

The HR20 is awesome.

That 5lnb dish is sweet.

I love D*

I can't wait for another update.

Updates rock.

Encryption is amazing.

Yes sir I'll have another...

jonaswan2

12-12-06, 04:16 PM

Uh, thanks, I appreciate at least one person seeing this with anything other than geek eyes. Not name calling just saying...anyway, seems to be many conflicting ideas on how this thing works. No need to continue. Since I'm the one who started this thread and I basically know now that no one really knows, then cool. No biggie, I don't really care that much. Just a question. Maybe I'll just keep my post to the following...

The HR20 is awesome.

That 5lnb dish is sweet.

I love D*

I can't wait for another update.

Updates rock.

Encryption is amazing.

Yes sir I'll have another...

Your right that not to many people know the what processor the HR20 uses, not even Earl.

iacas

12-12-06, 05:10 PM

Actually it does encryption, I'm pretty sure. I'm sure they based the current methodology off of the Tivo "style" of doing things, which involved having a crypto chip onboard each unit with a unique key, so that recordings made on that unit can only be played back on that unit.

I'm saying that I don't think they do constant or "on the fly" decryption. Each recorded file likely contains an encrypted key somewhere in the header. Once it's decrypted - a one-time operation - the file plays. Likewise, it probably only encrypts recorded files just the one time. It would make no sense to encrypt the entire stream of data and have the box encrypting and decrypting on the fly.

The original poster (the "ignore" list comes in handy, incidentally) was talking about encrypting and decrypting as if it happened on the fly. The only thing that (likely) happens on the fly is decompression, because clearly that's a "must" for playback. Almost no form of encrypted data requires or employs on-the-fly encryption and decryption. Even the iTunes media stuff with DRM is encrypted in only a single location, in a "wrapper" element. It doesn't decrypt on the fly either.

lguvenoz

12-12-06, 06:10 PM

I'm saying that I don't think they do constant or "on the fly" decryption. Each recorded file likely contains an encrypted key somewhere in the header. Once it's decrypted - a one-time operation - the file plays. Likewise, it probably only encrypts recorded files just the one time. It would make no sense to encrypt the entire stream of data and have the box encrypting and decrypting on the fly.

The original poster (the "ignore" list comes in handy, incidentally) was talking about encrypting and decrypting as if it happened on the fly. The only thing that (likely) happens on the fly is decompression, because clearly that's a "must" for playback. Almost no form of encrypted data requires or employs on-the-fly encryption and decryption. Even the iTunes media stuff with DRM is encrypted in only a single location, in a "wrapper" element. It doesn't decrypt on the fly either.

I would be curious as to whether or not D* does this or not. I am much more inclined to believe they went the same route as the new PCs are heading. Add on a single chip and let it handle the encryption and decryption as the data flows into and out of the drive. It's really not a big deal, and is the only thing that makes sense if they want to avoid people popping the drive out and cloning the digital content.

I think it would be interesting if there was anyone out the with a list of the various processing chips that are used within the HR20: CPU, GPU, MPEG2/4, etc. This would be very indicative of the overall processing capability of the unit, and would even point the finger at things that might slow the machine down sometimes (my gut feeling is IO or processing bottlenecks with IO being the most likely culprit).

iacas

12-13-06, 09:14 AM

I would be curious as to whether or not D* does this or not. I am much more inclined to believe they went the same route as the new PCs are heading. Add on a single chip and let it handle the encryption and decryption as the data flows into and out of the drive. It's really not a big deal, and is the only thing that makes sense if they want to avoid people popping the drive out and cloning the digital content.

They may have a chip to do it, yes, but in that case it still wouldn't have anything to do with the OP's inability to record two shows nor would it have anything to do with the "main CPU's processor speed or equivalent."

If the HR20 writes the data it receives, even if it's "pre-encrypted" but still inside the "box," then it's still just a write. Where the decryption - if there is "on the fly" decryption which I still doubt - because of the added decompression step, I wonder where the decryption would fall in that line of events: read-decrypt-decompress or read-decompress-decrypt? Seems to me if encrypted data is stored it'd be the first, but again, that's iff (if and only if) on-the-fly encryption is used. I still bet that's a "no." Though, since on-the-fly encryption is harder, it could explain why DTV has so much trouble with this box. :)

I think it would be interesting if there was anyone out the with a list of the various processing chips that are used within the HR20: CPU, GPU, MPEG2/4, etc. This would be very indicative of the overall processing capability of the unit, and would even point the finger at things that might slow the machine down sometimes (my gut feeling is IO or processing bottlenecks with IO being the most likely culprit).

I/O on any HD-capable machine is indeed a typical bottleneck. I too would wonder what the exact specs of most of the chips are and the flow or order through those chips.

banningview

12-13-06, 12:54 PM

They may have a chip to do it, yes, but in that case it still wouldn't have anything to do with the OP's inability to record two shows nor would it have anything to do with the "main CPU's processor speed or equivalent."

If the HR20 writes the data it receives, even if it's "pre-encrypted" but still inside the "box," then it's still just a write. Where the decryption - if there is "on the fly" decryption which I still doubt - because of the added decompression step, I wonder where the decryption would fall in that line of events: read-decrypt-decompress or read-decompress-decrypt? Seems to me if encrypted data is stored it'd be the first, but again, that's iff (if and only if) on-the-fly encryption is used. I still bet that's a "no." Though, since on-the-fly encryption is harder, it could explain why DTV has so much trouble with this box. :)

I/O on any HD-capable machine is indeed a typical bottleneck. I too would wonder what the exact specs of most of the chips are and the flow or order through those chips.

So you are saying that it is possible that some encryption is happening in the box and that could be a bottleneck slowing down the machine. Enough to cause the box to have audio dropouts? What about sending info down on FOX? Enough to cause a problem?

iacas

12-13-06, 01:20 PM

So you are saying that it is possible that some encryption is happening in the box and that could be a bottleneck slowing down the machine. Enough to cause the box to have audio dropouts? What about sending info down on FOX? Enough to cause a problem?

No, I'm not saying that at all. If there is an encryption/decryption chip, failure of that chip would likely lead to NO playback at all. If there is such a chip, too, it would perform at a speed that far exceeded the required rate. You wouldn't get a "slow" chip that sometimes dropped audio and sometimes not.

lguvenoz

12-13-06, 01:33 PM

No, I'm not saying that at all. If there is an encryption/decryption chip, failure of that chip would likely lead to NO playback at all. If there is such a chip, too, it would perform at a speed that far exceeded the required rate. You wouldn't get a "slow" chip that sometimes dropped audio and sometimes not.

I would concur with this. If they use a hardware device to do on-the-fly encryption/decryption of data as it moves to and from the hard disk, I would never expect it to cause these types of issues.

banningview

12-13-06, 01:39 PM

I would concur with this. If they use a hardware device to do on-the-fly encryption/decryption of data as it moves to and from the hard disk, I would never expect it to cause these types of issues.

Hmmm...I've always noticed that there is a lag from the DTV channels to the OTA channels. Could it be that the information is being buffered on the disk and the management of that data is slowing things down, or possibly causing it to hiccup in spots.

jonaswan2

12-13-06, 02:57 PM

banningview

12-13-06, 03:11 PM

The access card decrypts the TV info, it's really not that complicated. And the box doesn't encrypt the data, the data is already encrypted and stays encrypted while it's on the disk. The access card just has to decrypt how ever it decrypts and the box does all the rest.

O.K. that's something I haven't heard, but actually makes sense. I thought the access card for simply for that...access. Once the card is in I thought it just allows activation of the box.

So I'm guessing now that the real problem is that DTV is having a hard time sending a decent mpeg4 signal for some of the feeds or maybe the way the local stations are sending their OTA is messing them up.

My box has been working perfectly since day one, not a single restart except on purpose to see if there is a software update.

The HR20 is awesome.

lguvenoz

12-13-06, 04:35 PM

P Smith

12-13-06, 08:33 PM

iacas

12-13-06, 08:38 PM

Total mess.:rolleyes:

A card do not decrypt video/audio - for that it need much more memory and power. That IS a task for STB itself.

The access card basically contains a number. When you change your service, D* sends a programming change down to thousands of receivers. If the number matches, your box latches on to the programming change and suddenly you start seeing the Playboy TV (or whatever) that you just subscribed to.

It does no encryption. It would need a processor of some kind to do that.

The box itself also contains a number (even the serial number would work as a seed). You can switch access cards and, once the new # is provisioned, your recorded shows still play. You can also put the old hard drive and the old card in a new box and the shows won't play.

IF the HR20 encrypts and decrypts on the fly, it's most likely to be a separate chip, and NOT one on the access card.

jonaswan2

12-13-06, 09:16 PM

The access card basically contains a number. When you change your service, D* sends a programming change down to thousands of receivers. If the number matches, your box latches on to the programming change and suddenly you start seeing the Playboy TV (or whatever) that you just subscribed to.

It does no encryption. It would need a processor of some kind to do that.

The box itself also contains a number (even the serial number would work as a seed). You can switch access cards and, once the new # is provisioned, your recorded shows still play. You can also put the old hard drive and the old card in a new box and the shows won't play.

IF the HR20 encrypts and decrypts on the fly, it's most likely to be a separate chip, and NOT one on the access card.

It kind of seems more complicated than that, but I think you've just described VideoGuard. The content is saved with that "number" and the access card cracks it (I should of explained it like that in the first place, much more accurate). There is no decryption or encryption going on in that box as far as we know. If there was, there would probably be a SVP inside the box, since that's NDS's DRM of choice (but not necessarily DirecTV really).

banningview

12-13-06, 09:42 PM

jonaswan2

12-14-06, 05:21 AM

Frustrating isn't it. I mean, it's not what you say, it's what he says, but he really doesn't know anything just as long as what you say is wrong or so he says...

Does anyone REALLY know what is going on in the box...

After reading these posts it seems that from my post to the OP, no one really knows, it's all just conjectures and guesses. As long as he is right and everyone else is wrong...:lol: :lol: :lol:

Only Earl knows some things that go power the box and some features that we don't know, but he probably knows all about. No one but DirecTV and NDS know how the CA works and it's best that it stay that way :D.

I don't think that any one, except some DirecTV engineers and software devs, knows anything about the inside of the box, how it works, and how all of the various parts work together, we can only guess and assume based off of the information that were drip fed. That should probably also stay the way it is, it's not DirecTV recievers have open source software.

P Smith

12-14-06, 09:39 AM

The knowledge is out there, there is no mistery what STB and smart card doing.
Some keys, very specific routines and key permutations algos unknown, but it totally irrelevant for the discussion.

So, briefly: the card getting encrypted key from stream thru STB, then check if the channel present in the card's internal database, then send the key back to STB, where the key used for decryption of video/audio.
That's essential part of whole process of viewing subsribed channel.

Oh, BTW the card have CPU,RAM,ROM,EEPROM and special part known as ASIC.

iacas

12-14-06, 09:45 AM

Oh, BTW the card have CPU,RAM,ROM,EEPROM and special part known as ASIC.

I really, really doubt that the DirecTV access card contains a CPU, unless you're stretching the definition to include the very fringe. What are your sources?

LameLefty

12-14-06, 09:53 AM

I really, really doubt that the DirecTV access card contains a CPU, unless you're stretching the definition to include the very fringe. What are your sources?

Not many sources about D* smartcards around anymore - between the DMCA making such discussions essentially illegal and D* suing every individual in the U.S. who buys smartcards and readers/programmers for signal theft and piracy, I say "Good luck!" finding much hard information outside of corporate design labs protected by NDAs and serious threats.

jonaswan2

12-14-06, 09:59 AM

Not many sources about D* smartcards around anymore - between the DMCA making such discussions essentially illegal and D* suing every individual in the U.S. who buys smartcards and readers/programmers for signal theft and piracy, I say "Good luck!" finding much hard information outside of corporate design labs protected by NDAs and serious threats.

Ummm, the US patent office? It says in one of those NDS patents clearly that the smart card has a semiconductor chip inside. Proof enough to me! :D There is also some patent numbers on your access cards you can search for.