SDV FAQ

[mikeyts]

Originally Posted by bicker

I wonder what the significance is of the specification just being released. Does this provide any more insight into the timetable (i.e., "not soon")?

Actually, it's a good sign. One of their barriers to releasing the spec was doing an IP search to make sure that their design didn't cross any existing patents. So that's done and people can manufacture to the spec without fear of surprise litigation.

[morac]

I notice that there is a special message sent by the resolver in the case where a channel is requested, but not by the user:

Originally Posted by page 23

Type of tune request.
0x0 – Live full screen video (no HDD recording)
0x1 – Live full screen video (HDD recording)
0x2 – Live PIP or POP video (no HDD recording)
0x3 – Live PIP or POP video (HDD recording)
0x4 – Recording only
0x5 – Inactive
0x6 – Speculative Recording
0x7 – Reserved

In the resolve_tuning_req() APDU, the UDCP SHALL only report the tuner_use_status values associated with
HDD Recording (0x1, 0x3 or 0x4) when a user-initiated, scheduled or timed recording is in effect that will always
be followed by an udcp_status_update() APDU that reports 0x0, 0x2, or 0x5, 0x6 at the end of the recording.

In the resolve_tuning_req() APDU, the UDCP SHALL report tuner_use_status values associated with opportunistic
or speculative recording that was not user-initiated as 0x6.

I'm assuming a value of 6 would be used for suggestions. This makes sense since requests can then be given priority depending on what it being done. For example, I'm assuming a "speculative recording" (ie: Suggestion) would be given the lowest priority.

Also see section 10.2 for examples of confirmation windows to confirm you are watching the channel.

[bicker]

Originally Posted by jlb

Nice circular reference. Efficient.

That's not a circular reference. It is a forward reference. It is not uncommon for one word to say its definition is the same as the definition for another word, and then have that other word mention that there is another word that is synonymous.

[husky55]

The concern I have is there is no clear schedule of the SDV dongle rollout for Tivo from the cable cos. Some vague date like 2H 2008. Will our Tivos become dinosaurs in this new SDV HD world?

[mikeyts]

Originally Posted by husky55

The concern I have is there is no clear schedule of the SDV dongle rollout for Tivo from the cable cos. Some vague date like 2H 2008. Will our Tivos become dinosaurs in this new SDV HD world?

And what reason would they have to publish a precise schedule to the public? To make you feel better? I don't think so. That type of crap will get you sued for missing the date by a month or two. In the engineering world you don't issue completion guarantees unless you have to. by the end of the 2nd half of calendar '08 is good enough.

[moyekj]

This thought occurred to me in the unencrypted QAM thread...

I wonder if the tuning resolver could be used in units without cablecard? It may offer channel mapping capabilities of it's own such that unencrypted channels would be properly mapped to the cable company channel #s and solve the unencrypted QAM without cablecard problem.

[pmiranda]

Originally Posted by moyekj

I wonder if the tuning resolver could be used in units without cablecard?

It kinda sounds like it from the spec, although in a twist of fate, the FCC rules might actually require a cablecard in the TR if it can operate without one in the TiVo :-)

[morac]

Apparently CableLabs just got the tools to test the tuning resolver. This means they haven't had a chance to test it yet. I'm also wondering how ready it will actually be if they do deploy it by 2nd quarter. I'm wondering what would be worse, not having the tuning resolver or having it and finding that it flakes out half the time because they didn't have a chance to test it fully before deployment.

[mikeyts]

Originally Posted by morac

Apparently CableLabs just got the tools to test the tuning resolver. This means they haven't had a chance to test it yet. I'm also wondering how ready it will actually be if they do deploy it by 2nd quarter. I'm wondering what would be worse, not having the tuning resolver or having it and finding that it flakes out half the time because they didn't have a chance to test it fully before deployment.

As an engineer with 30 years experience (12 of them in development of large scale networking devices), I think that they've given themselves plenty of time to get this thing to market. It depends on how many resources the various players (Motorola, SA, TiVo, the cable providers, etc, etc) have devoted to the effort--with enough people and dollars it could conceivably be done in significantly less time. This is, conceptually, not a very complicated project--I've been on teams that got out new products with many hundreds of thousands of lines of integrated code, much of it newly written, in less time.

Chill, people, and stop fretting about it .

[mercurial]

Originally Posted by morac

Apparently CableLabs just got the tools to test the tuning resolver. This means they haven't had a chance to test it yet. I'm also wondering how ready it will actually be if they do deploy it by 2nd quarter. I'm wondering what would be worse, not having the tuning resolver or having it and finding that it flakes out half the time because they didn't have a chance to test it fully before deployment.

Or dealing with the compression TWC is using around here that's really killing a bunch of SD channels right now. I'd almost not get some channels than see some of the painful artifacts.

[CharlesH]

I suspect that the resolver will pretty much be a stripped-down set-top box, rather than something developed from scratch. Remove the audio and video hardware, and add the software to talk the host device. That functionality is pretty much the only new software that has to be written; all the other stuff, such talking the SDV protocol to the head end, pulling channel maps off the cable, and such, is already there. For that reason, there is technically no reason that it also couldn't handle VOD and PPV for the host, but it seems that OCAP in the host is the way they want to go.

[ah30k]

It was already stated to be the same form factor as the DCT-700

[mikeyts]

Originally Posted by ah30k

It was already stated to be the same form factor as the DCT-700

More like rumored. Motorola hasn't officially said anything about it at all. What we "know" about it we "know" from the blog of someone with "inside Motorola contacts". Take it all with a big bucket of salt.

[lrhorer]

Originally Posted by ah30k

It was already stated to be the same form factor as the DCT-700

If it is in fact true, it's only true for Motorola systems. Most of us are not on Motorola systems.

[lrhorer]

Originally Posted by mikeyts

As an engineer with 30 years experience (12 of them in development of large scale networking devices), I think that they've given themselves plenty of time to get this thing to market.

I don't think that's really the issue. There's not going to be any large scale re-tooling, and the functional blocks already exist, so no, no motivated engineering team should have a problem getting this out, unless of course infighting among the various UDCP manufacturers caused a big delay. Between the fact the TiVo is virtually the only device which will be able to make use of the TR and the fact the standard has now been published, this problem is pretty much moot.

Originally Posted by mikeyts

It depends on how many resources the various players (Motorola, SA, TiVo, the cable providers, etc, etc) have devoted to the effort

THAT's the issue, and it boils down to how motivated the marketing authorities in the respective companies are to induce the accounting executives to approve resources for the engineering teams to produce the product. In some companies, that level may be less than zero, with one or more members of the senior executive team being actively opposed to its development. We all know what "The check's in the mail" means.

Originally Posted by mikeyts

--with enough people and dollars it could conceivably be done in significantly less time.

Yeah, but with any significant amount of opposition within the company's power structure, the company might well never spend a single penny to develop the dongle. Without an FCC mandate, there won't be anything forcing the company to bother, and since most markets don't have competing CATV systems, the fact the S3 and other UDCP devices are pretty much locked into using a CATV provider implies there can be real marketing arguments for not wanting to develop the dongle in the first place.

[ah30k]

Originally Posted by ah30k

It was already stated to be the same form factor as the DCT-700

Originally Posted by lrhorer

If it is in fact true, it's only true for Motorola systems. Most of us are not on Motorola systems.

True, my frame of reference is Mot so I tend to only think that way. How do you figure 'most of us are not on Mot systems'? I was under the impression Mot had a rather significant market share.

[lrhorer]

Originally Posted by mikeyts

And what reason would they have to publish a precise schedule to the public? To make you feel better? I don't think so.

Well, some companies do. One of our major vendors, Nortel , for example is usually very good about forecasting release dates. They usually hit them, too. It's one reason we're loyal to them. They keep us informed of their development plans and meet their commitments. It's not breaking any law not to do so, however.

[lrhorer]

Originally Posted by ah30k

True, my frame of reference is Mot so I tend to only think that way. How do you figure 'most of us are not on Mot systems'? I was under the impression Mot had a rather significant market share.

A significant market share is not most. Less than half of all subscribers are on Motorola systems, so by definition most are not on Motorola systems. It's been some time since I looked at the stats, but there may not be a single majority SDV vendor. If there is, it's Scientific Atlanta.

[lrhorer]

Originally Posted by CharlesH

I suspect that the resolver will pretty much be a stripped-down set-top box, rather than something developed from scratch.

That's a reasonable way to handle it. So is taking a DOCSIS modem and adding a USB 2.0 port and some software.

Originally Posted by CharlesH

pulling channel maps off the cable, and such, is already there. For that reason, there is technically no reason that it also couldn't handle VOD and PPV for the host, but it seems that OCAP in the host is the way they want to go.

The host software required to interface with the dongle is minimal. VOD, PPV, and other non-primitive 2-way functions are vastly more sophisticated, not to mention being highly proprietary to each CATV system. There are only a handful of SDV protocols and hardware for handling scheduled programs. The number of proprietary interactive systems is vast.

[ah30k]

Originally Posted by lrhorer

A significant market share is not most. Less than half of all subscribers are on Motorola systems, so by definition most are not on Motorola systems.

I guess that is where I must have been off. I thought Mot had more systems share than SA. Where do you get your data?

Yes, SA is ahead on SDV but not for long.

[mikeyts]

Originally Posted by lrhorer

I don't think that's really the issue. There's not going to be any large scale re-tooling, and the functional blocks already exist, so no, no motivated engineering team should have a problem getting this out, unless of course infighting among the various UDCP manufacturers caused a big delay. Between the fact the TiVo is virtually the only device which will be able to make use of the TR and the fact the standard has now been published, this problem is pretty much moot.

TiVo is virtually the only device that can use this thing now, but conceivably new, low-end DCR devices could be designed to be able to use it from the beginning. I think that's the suggestion. "No, we don't need to spend a ton of time and money completing the definition of and providing compliance to the CEA's proposed 'Digital Cable Ready Plus' standard--with this simply little device, which we've already defined and prepared for distribution with the aid of a few key CEA members, the OEMs can make future low-end products which can tune SDV services without having to feature an expensive OCAP platform. (It doesn't address IPPV and VOD like DCR+ does, but of course SDV's the important thing)".

THAT's the issue, and it boils down to how motivated the marketing authorities in the respective companies are to induce the accounting executives to approve resources for the engineering teams to produce the product. In some companies, that level may be less than zero, with one or more members of the senior executive team being actively opposed to its development. We all know what "The check's in the mail" means.

"The check's in the mail", will only keep your landlord from kickin' your ass out for a very short time. They're trying to deliver on this almost before the FCC can require it in order to make it the solution of choice, since it'll be a solution that's already available or very close to being available. I still think that the FCC will need to require it, forcing the smaller providers to stock and provide the things.

[mikeyts]

Originally Posted by lrhorer

That's a reasonable way to handle it. So is taking a DOCSIS modem and adding a USB 2.0 port and some software.

Every DOCSIS modem I've seen for years has had a functioning USB 2.0 port. Cable modems are a lot closer to being physically a match for what's needed than any tuning STB, which have unnecessary graphics chips and audio chips and IR receivers and way too much memory. If Moto's using the DCT-700 as a base for development instead of a cable modem, it's because the Tuning Resolver is being engineered by the tuner STB team and they're a separate group from the networking equipment team who are already familiar with writing code for the DCT-700.

Originally Posted by lrhorer

Well, some companies do. One of our major vendors, Nortel , for example is usually very good about forecasting release dates. They usually hit them, too. It's one reason we're loyal to them. They keep us informed of their development plans and meet their commitments. It's not breaking any law not to do so, however.

Oh, I am absolutely certain that the cable equipment OEMs have detailed internal schedules which feature versions of the Tuning Resolver as a deliverable, allocating resources to the production of it. You can't run a company on vague ideas of when products will be ready. They know exactly when they expect it to be done and refine that date all the time. I'll bet that they share that estimate with the cable providers, but giving that exact date to the public is another matter. You don't do that unless you have to and they don't have to. It might give us all the warm fuzzies, but there's no upside for the OEMs.

[CharlesH]

Originally Posted by lrhorer

The host software required to interface with the dongle is minimal. VOD, PPV, and other non-primitive 2-way functions are vastly more sophisticated, not to mention being highly proprietary to each CATV system. There are only a handful of SDV protocols and hardware for handling scheduled programs. The number of proprietary interactive systems is vast.

I was assuming that the STB already had the proprietary software to handle VOD and PPV, and what would have to be done is replace the GUI for those functions by a standardized API over USB.

[snowbunny]

My older S2 (single tuner) lifetime TiVo is having serious trouble with its hard drive (upgraded hard drive, no problems until now). With the dongle repeatedly slipping schedule, I have no idea what to do -- try to have the S2 repaired? Buy a new S2 (but what about my lifetime)? Buy a new S3/HD with lifetime, which will mean losing many SDV channels? I'm getting repeated complaints from the kids and my visiting relatives about not being able to receive channels on the S3 -- it's simply that bad with as many SDV channels as we have (and we're not trying to even watch HD yet).

I'd really hoped the dongle would be on time as reported last fall... apparently not. We do not have any other competitor in the area for cable so are at the mercy of TWC.

Would you pay to have the S2 repaired if possible, and if not, which S2 would you buy?

[moyekj]

Originally Posted by snowbunny

With the dongle repeatedly slipping schedule....

I'd really hoped the dongle would be on time as reported last fall... apparently not.

Please cite where the dongle (AKA tuning resolver) was PROMISED by anyone to be available last fall? AFAIK the only information available was a somewhat vague Q2 2008 that Motorola MAY have something ready to go. Plus I've seen no information from Scientific Atlanta (Cisco) on what their solution may look like or when it may be available.

[mikeyts]

There has been no published "schedule" to be "slipped". The only thing we have is an estimate that at least some version of it should be being distributed to customers by the end of the second quarter of calendar '08. Unless every company which plans to manufacture a version of the Tuning Resolver has chosen to assign it to a team of 2 or 3 of their least talented engineers, that was a pretty generous estimate. Until we get some official notification that no version of it will be ready before the end of June, there has been no schedule "slip".

If I were you, snowbunny, I might consider leasing a TWC DVR in the interrim. (I'm actually considering leasing a non-DVR STB to get access to all the pay and subscription HD VOD that's starting to become available). Where are you located that SDV has become such a big problem (consider placing your location in your forum user profile).

[jrm01]

Originally Posted by moyekj

Please cite where the dongle (AKA tuning resolver) was PROMISED by anyone to be available last fall? AFAIK the only information available was a somewhat vague Q2 2008 that Motorola MAY have something ready to go. Plus I've seen no information from Scientific Atlanta (Cisco) on what their solution may look like or when it may be available.

Does the Tuner Resolver have to be specific for SA vs. Motorola? I thought that it would only be interacting with the host (Tivo) and the neighborhood hub (which I thought was third party, not Mot or SA). Since it would be independant from cablecard and headend gear does it matter?

[mikeyts]

Originally Posted by jrm01

Does the Tuner Resolver have to be specific for SA vs. Motorola? I thought that it would only be interacting with the host (Tivo) and the neighborhood hub (which I thought was third party, not Mot or SA). Since it would be independant from cablecard and headend gear does it matter?

It's gonna have to speak the network SDV protocol, and whereas some of that stuff is by third parties, its still proprietary and there are multiple versions in use.

[classicsat]

In a nut shell, it is SA or Motorola hardware provided by your cable provider. By it being from the respective manufacturer of their existing cable gear, it will be somewhat easy for a provider to integrate into their system.

[snowbunny]

Originally Posted by moyekj

Please cite where the dongle (AKA tuning resolver) was PROMISED by anyone to be available last fall? AFAIK the only information available was a somewhat vague Q2 2008 that Motorola MAY have something ready to go. Plus I've seen no information from Scientific Atlanta (Cisco) on what their solution may look like or when it may be available.

My original quote was that the last estimate I saw for dongle availability was *from* last fall, and the projection was Q1 2008.

I'm in Austin so am hard hit by SDV. I want to scream when my mother asks me why one TV can't tune in her favourite channels.... but the other one (with the dying TiVo) can.... I keep telling her that no, it's not broken, it's a technical incompatibility that will be fixed but I don't know when. Lather, rinse, repeat.