[louis]

Hi,

i just had a look at the XBMC Roadmap, and there is no PVR functionality planned anymore in XBMC 12.0?! As far as i remember it was planned to integrate PVR into 12.0...now it is only mentioned in "Milestone: future/pending".

Did i miss something? Or is this just an mistake? Thanks for clarification...

Ciao Louis

[jmarshall]

There is no mistake. Features get in when they're ready. PVR is not ready and until the API is stabilised will be staying where it is.

It's always better to have things NOT on the roadmap so that we ensure things are achievable.

A lot of the stuff on there is still likely to not be done. I have no idea who keeps adding stuff to it :p

Cheers,
Jonathan

[louis]

Ok Jonathan, thanks for clarification...

Cheers Louis

[Jezz_X]

And they just plain get it wrong sometimes too

A new Confluence "Light" for embedded devices.

Its actually for all devices not just embedded

[bmn001]

So...PVR moved out of Frodo. I understand how these things go, and am not complaining. Thanks for taking the time to do it right.

But I have to say this: "NOOOOOOOOOOOO!"

[KRA77]

Seen this so many times now, it's no surprise. Before 10.5 (wich ended as 10.12?) there was talk that pvr would come in next release.

Still though, we got a fairly ok working pvr branch, even though it's not official.

[MakkaD]

The idea i had since the beginning is that nobody have the minimum clue on what is needed to "stabilise the API".
Actually it looks like there is no effort by the developer on PVR or the interest to spend time on it.
What are the differences since 2 years ago?
What work has been done since then to "stabilise the APIs"?
Which are the showstopper missing features that do not let the PVR branch get into official build?
I have not seen any wiky or documentation or analisys on all these aspect so no surprise the API are not stable and PVR is not yet "ready" for prime time.
The only way to introduce the PVR functionality is to go through these steps:
- PLANNING/FEATURE REQUIREMENTS: Feature definitions to make clear which features are "needed" and what "can wait" future version.
- DEFINITION: Define and stabilise the API that will be available to access the features
- DEVELOP: Start to produce the code needed to fulfill the requirements (creating an OFFCIAL branch?).
- INTEGRATE: Start to introduce the changes in official (trunk) code starting from the core changes and ending with the API. Introducing these features will make the code UNSTABLE which will stabilise with time and bug fixing
- BUG FIXING: To stabilise the code
These steps are to be reiterated at each new feature to introduce.
All these step have to be OFFICIALLY supported, the first two do not even need to change a line of code but without that the stabilisation of a feature is just impossible.
Now the branch is just something more than a proof of concept, there is no planning at all, or at least i have not found any, so nobody have an idea on what is needed and how to do it.
All the work is done by individuals, each one with his own vision, obviously, on what is needed and what can wait.
An example of that is the VNSI/XVDR debate which ended with the XVDR addon not to be supported or included in semi-official XBMC PVR builds loosing the functionalityes it have.
It is not a fault of Pipelka becouse he just "did what he wanted to", and it is not a fault of Dushmaniac (or XBMC developers in general) becouse he could not support it, either way we do not have an official advanced VDR addon (VNSI is all but advanced) and probably will never have it.
Dushmaniac is doing a lot of work but it looks like he is "alone" and not supported by the rest of the developers comunity.
I know that developers are not pyed in any way to work on a particular feature or piece of code so each developer work on "what he likes to", but it is nontheless true that PVR is THE MISSING FEATURE of XBMC.
So, anyone who loves XBNC as I do, should be happy to work on that and probably WILL, looking at the number of people and work present in this section of the forum.
What is missing is clear idea of what to do and how to do it in order to have PVR functionalityes in XBMC.
After my "rant", i really enjoy the slick interface of XBMC, its features and the way content are presented BUT, i am thinking of moving away from all these goodies in order to have some feature i really need:
- Live TV
- Timeshift
- Advanced Recording Timer
- Good EPG support
Ok, the interface will be not so beatifull, the content will not be so rich of linked data etc., but it will do what my htpc is meant to do: be an easy and complete interface to my content (live or stored).

Makka

[Jimmer]

After my "rant", i really enjoy the slick interface of XBMC, its features and the way content are presented BUT, i am thinking of moving away from all these goodies in order to have some feature i really need:
- Live TV
- Timeshift
- Advanced Recording Timer
- Good EPG support
Ok, the interface will be not so beatifull, the content will not be so rich of linked data etc., but it will do what my htpc is meant to do: be an easy and complete interface to my content (live or stored).

Makka

I'm using the mythtv addon and it has all that stuff pretty much (I'm not really sure what qualifies as 'Advanced Recording Timer' so maybe it doesn't have that...).

To go back to your earlier point, Dushmaniac is developing the PVR stuff and setting the API. As well as the main PVR code, he's responsible for the TVHeadend addon. Other developers fork his code and maintain their respective addons. Stuff gets pushed back to Dushmaniac's repo when it's ready. By looking at the pull requests on Dushmaniac's repo it seems clear that there is a lot of back-and-forth between Dushmaniac and the various add-on devs. Everyone of them seems to be aware that Dushmaniac's about to do a major re-write and are waiting for this to be finished before they make changes to their code which will become obseleted by the impending changes.

It's too bad about the VDR thing... but people need to co-ordinate with the main man on this one..... period. Anything else is just madness.

Finally, PVR isn't really missing. Just pull down the repo you're interested in and compile it. I use it on all my rigs and it works just fine. Else, just grab an "official" version from pulse-eight (or wherever they come from nowadays) and have at it

Just my 2p

[audimax]

It's always better to have things NOT on the roadmap so that we ensure things are achievable.

And yet it's been in there for over two years ....

And they just plain get it wrong sometimes too

You have got to be kidding.
It's been in there for over two years and now it's suddenly somebody put it there by accident?

I too know, you guys do all the work in your free time and i really respect that. You have done a great job and i really love XBMC.
But right now i feel like PVR as a XBMC Feature has been promised at least one year ago.

I can only support everything MakkaD said. He just got it right.

There is only one question left to every XBMC Team Member who cares to answer:
Will there be a PVR Integration in XBMC or are you guys trying to stall an wait for a future (maybe in 10 to 20 years) when there is no more live tv but only streaming?

a "slightly" disappointed
max

[Jimmer]

It's always better to have things NOT on the roadmap so that we ensure things are achievable.

And yet it's been in there for over two years ....

And they just plain get it wrong sometimes too

You have got to be kidding.
It's been in there for over two years and now it's suddenly somebody put it there by accident?

He's talking about the reference to his skin on the roadmap, not PVR. At least quote the guy in context!

[DDDamian]

Grouchiest thread I've seen in a while......sheessh

[Jimmer]

Grouchiest thread I've seen in a while......sheessh

I tried.....

[erichzann]

These changes in the road map really don't worry me too much.

I have been running MePo for a number of years on our main living room HTPC and am in the process of moving over to XBMC PVR (NYXBoard Hybrid remote is in the post!!!)

I've been using the PVR builds on my own machine for ages and already there are a couple of features that MePo doesn't support (yet):

- 'Coloured' EPG guide (probably going to be in MePo 1.3)

- Filter by channel name

- Really fast channel changes (seems quicker than MePo)

- Group Management built into the GUI

Yes, the advanced record features are missing but I generally only record films so this isn't a problem.

Also, time-shifting will be great but being able to pause the buffer is enough to make a phonecall or grab a drink, so it's not a problem.

Also, just because it isn't going to be merged for the next release doesn't mean the development will stop.

I've been using the MePo Sky EPG SVN build for the whole time I've used MePo and this hasn't been a problem at all.

[sho]

The roadmap is in no way a promise of features, it's written by people like me that couldn't code their way out of a paperbag.
Also things change, live with it. (And btw this is a free world, use whatever software you feel is right for you).

[xbmcuser2000]

mmm...

I think (and hope) that the XBMC Team are joking and they will present PVR merge in May Cycle.

From conclusions on April Cycle Post:

...keep an eye out for the May Cycle. Our developers are working extra hard to hopefully include one of the most requested feature additions of the past two years.

And if I am wrong (), thanks anyway for your great work.