[Vertigo]

Merging a HTPC and AV amp

<rant>
Ive been using (budget) HTPCs on and off for over 10 years now. Currently Im using a recycled PC running XBMC connected straight to my TV through HDMI. I also use a proprietary cable box from my local cable provider. Since neither of these devices truly supports CEC (other than turning on/off), its a bit of a remote control hassle. I contemplated buying a Harmony to fix the problem, but I know other household members will use the specific remotes and cause the harmony to get “unsyched”. As I see it, Harmonys are a decent attempt to fix a problem that shouldnt exist. Anyway, I could get around with 2 remotes, since the cable box remote also allows limited control over the TV.

Until I bought a new tv; the built in speakers where absolutely aweful (shame on you Phillips), so I decided to attach a decent audio system based on a AV receiver. Thats were the problem truly began. Ive bumped my head in to every conceivable problem, such as the HTPC dropping HDMI audio when switching tv inputs, forcing me to use SPDIF, requiring yet another remote to control volume. Problems with audio/video sync. Problems with the receiver not supporting certain audio codecs (like AAC 5,1), problems with CEC not allowing control when the tv was turned off, etc, etc.

Now I have a degree in ICT and generally Im a gadget freak; if Im finding it hard to find the right equipment to buy, set this all up properly and use it, how will joe 6 pack ever manage?
He wont.
</rant>

AV suppliers know this and I noticed that recent AV receivers and Tvs alike are increasingly becoming “smart”. They have network connection and allow streaming audio, or even limited video. While thats great in theory, in reality its like going back 20 years in time compared to using XBMC. You pay $2000 for a high end AV receiver and you end up with a media interface that looks no better than a DOS file manager back in 1980. The smart Tvs look slightly better, but unless you plan to stick with their built in sound, its not much of a complete solution either and moreover, their support for video codecs seems woeful and the experience is still a very far cry from XBMC.

That got me thinking; there has to be a better way.

Why not integrate an XBMC box with a decent AV receiver? There is ample room inside highend AVR's enclosures to house a raspberry pie to run a customized XBMC that takes control of everything. Use the XBMC interface to tune the radio, to switch between watching tv and movies or online movies or whatever other apps you want with ease. If you run it on Android you might even game on it. Just connect a TV, a network and you are done.. Even Joe 6 pack with a thick enough wallet could do it.

Now as long as Denon/Pioneer/Marantz etc are not stepping up to the plate, I wonder if this could be done by a kickstarter style project. The XBMC part should certainly be doable if you find the right people, but building, and integrating it in a decent sounding amplifier might not. Unless perhaps you use tripath. Not everyone seems to agree, but many audiophiles swear tripath amps sound as good as ultra high end tube amplifiers, yet they are dirt cheap, small and low power. Most seem to have relatively low output, but probably enough for most people. Also Ive not yet seen a 5,1 capable tripath amplifier, but you might just use one amp for every channel, which would also help boosting the output power. The company behind tripath seems bankrupt though, so i have no idea if these or similar chips are still available?

Another problem to solve would be the HDMI switching, and potentially upscaling, though I suspect daugher boards exist to allow just that?

Anyway, Im just floating this idea to see if people would be interested and whether or not it would be feasible. I certainly hope some day there will be a project like this on kickstarter. A true XBMC box that does everything.

[Livin]

In theory this is a nice idea... in reality, it will not be a viable business until the average AV consumer becomes much more tech savvy - maybe 10 years from now.

I actually worked for a very large computer tech company that looked into something similar to what you are suggesting, it is not a money making venture right now.
And... the aforementioned companies are not interested in making anything that is so open... They like their proprietary, does not play well with others systems.

For those of us using XBMC, and the like, we use the receiver as a back-end processing/amp device... not a front end, so for those who 'might' buy it, we are already happy with the solutions we have.

[Vertigo]

In theory this is a nice idea... in reality, it will not be a viable business until the average AV consumer becomes much more tech savvy - maybe 10 years from now.

The idea is actually to make it easier and put the richness of XBMC within reach of non techies.
For some perspective, after I set it up properly, my 77 year old mother now uses XBMC with relative ease.
But I dont know anyone besides myself who is able to configure all the machines (PC, tv, cable box, AVR, pulse-eight etc..). Certainly not my mother.

An AVR with a built-in (even if stripped down) XBMC node should be pretty much plug and play, not unlike what most modern network connected AVRs do, except with a real proper media interface and so much extra flexibility.

I actually worked for a very large computer tech company that looked into something similar to what you are suggesting, it is not a money making venture right now.

I dont know about that. I know plenty of people interested in a media center PC, but who are (rightfully) frightened by the complexity. These people now generally buy streaming media boxes, separate amps and Logitech Harmonies to cope with it all. Hardly ideal.

And... the aforementioned companies are not interested in making anything that is so open... They like their proprietary, does not play well with others systems.

Meh. They all support DLNA these days, and usually Airplay. That said, the point of integrating a HTPC in to an amplifier is that you wouldnt need as much interoperability. Its the same reason everyone is jumping on "smart tvs", so you wouldnt need another settop box or HTPC. Perhaps one day smart tvs will actually achieve that, but we are still very far from that today.

For those of us using XBMC, and the like, we use the receiver as a back-end processing/amp device... not a front end, so for those who 'might' buy it, we are already happy with the solutions we have.

I use XBMC, but Id gladly retire the PC running it if XBMC came integrated in to my next AVR or even TV. I dont know about you, but the whole reason I run XBMC on a PC is because nothing else exists that gives me the functionality. its not because I like yet another device in my rack that I have to setup, configure, maintain, make work with all the rest.

[Livin]

I understand your idea... many have had it - years before your post. No one has done it ... for many valid reasons.

If you think you know better than companies that do these thing for a living, for profit, then please do... I'm all for it.

DLNA is not 'open' like you are talking about... putting XBMC "inside" an AVR means making the core system (firmware) accessible... that will not happen. Anything else is simply a serial and/or network connection to have bi-directional communication... which already exists in upper-end receivers. I use my PC to control my AVR and have for years - I NEVER need to config/touch my receiver.

Anyway... go for it... and when you get it going let us know.

And, btw... I hope you can compete with the $300 AVRs which is where most purchases are in that segment. Or, you can go after the higher-end $800+ segment... but then get ready for the questions like... Who are you? Why would I buy or trust equipment from - who are you again? How many years do you have in the industry? What warranty do I get and how can I be sure you (who are you) will back it?

[Ned Scott]

Not a bad idea at all, and some very excellent points being made. I think the real reason no one has done it before is one part marketing and one part cost. The cost part is now so cheap, especially when you're already mass producing hardware, that even the marketing point doesn't have to be that important. You could have a minority of your customers who use such features, but not lose any money on it.

People like having things all-in-one when they can. Just because an idea wasn't marketed well doesn't mean it's a bad idea. Look at the media center concept itself. Think of how many people could be better served by an XBMC setup but are not using it or have not even heard of it.

[Vertigo]

Not a bad idea at all, and some very excellent points being made. I think the real reason no one has done it before is one part marketing and one part cost. The cost part is now so cheap, especially when you're already mass producing hardware, that even the marketing point doesn't have to be that important. You could have a minority of your customers who use such features, but not lose any money on it.

I can think of a million reasons why the big brands havent done this yet and probably wont for many years to come. For starters, they simply arent very good with software, its not their core business, and they are conservative by nature. That doesnt mean it wont happen; how long did it take after MP3 and DivX became commonplace on PC before the traditional home AV companies fully started supporting it? How long have we been buying/building media streamers and HTPCs before AV and PC companies started selling such boxes or integrating their functionality into AV appliances?

But while they waited, small companies like KiSS (remember them?) jumped in to the void. If I remember correctly, KiSS was actually founded to create DivX hardware players, but dont quote me on that. Once KiSS proved there was a market, Cisco bought them. Similar story with Boxee, who's devices ended up getting sold by D-Link.

Anyway, my point is just that Pioneer or Denon etc arent going to be the first to do this, but whoever does, has quite some challenges to overcome. Marketing and cost arent the biggest one I think; Integrating a full "PC" and software stack as complex as a full flown linux OS and XBMC in to a high end audio or video device is certainly in a different league from making a DivX player and requires a range of skill sets that few -if any- companies currently posses. But I dont think its undoable either. Especially not now that we have openelec and ARM SoCs. All we need is a team of people with the right skill sets working together.

As for the issue of open firmware mentioned by Livin; while XBMC itself is licensed under GPL, that doesnt mean they would need to open their entire firmware. Im not expert in OSS licencing, but XBMC supports binary plugings now, so the manufacturer could just create plugins to interface with their hardware. Also, linux is GPL software too, doesnt seem to prevent many from releasing linux based appliances.

Last point; look at the ouya console. While that is a project aimed at a very low cost device, its a template for what I would like to see happen with an XBMC appliance.

[Vertigo]

I understand your idea... many have had it - years before your post. No one has done it ... for many valid reasons.

I dont care if Im the first one to have the idea, but Its only very recently that it has become at all possible. It was never feasible to integrate a full blown x86 PC in to an AVR, and its pretty recent that you can instead use a cheap low power ARM SoC to run XBMC on, and a lightweight embedded OS like OpenELEC. No wonder no one has done it yet, all the essential ingredients are still in beta. Now would be a great time to get started on this.

If you think you know better than companies that do these thing for a living, for profit, then please do... I'm all for it.

Thats not the point. Those companies will go there eventually (if not with XBMC, then with similar home brew software), I have no doubt about that. See the post above, they are just slow to adopt to such changes. But that doesnt mean it cant be done or cant be profitable earlier.

DLNA is not 'open' like you are talking about... putting XBMC "inside" an AVR means making the core system (firmware) accessible... that will not happen. Anything else is simply a serial and/or network connection to have bi-directional communication... which already exists in upper-end receivers. I use my PC to control my AVR and have for years - I NEVER need to config/touch my receiver.

How many still have or want a separate device for playing DVD, streaming network audio, streaming movies, PVR, couch surfing? Once upon a time, you needed a device for each of those functions. Now most of that is getting integrated. Im just talking about taking the next step. Why have a PC to control your AVR if the AVR could become the PC (or vice versa) ? You lower complexity, you reduce cable clutter, you increase ease of use and potentially lower cost. Whats there to dislike?

And, btw... I hope you can compete with the $300 AVRs which is where most purchases are in that segment.

I agree, thats one of the tricky parts. If you integrate it into an AVR, you better make an AVR that doesnt sound like crap. How hard that is to do, I honestly have no idea, but I imagine its certainly non trivial, especially if you are not in that business already. I suggested the class-T amps to get you there "cheaply", but its certainly not my area of expertise. If all else fails, perhaps its possible to just buy or licence these components. That you will end up with something considerably more expensive than a mainstream AVR is quite okay In my opinion though. You would get a feature set that no AVR on the market would come close to, not even the highest end ones. Price wise you would be competing with standalone AVR + HTPC, and at least initially you could even get away charging a premium for the integration. You just need to make sure it sounds good or at least good enough.

Or, you can go after the higher-end $800+ segment... but then get ready for the questions like... Who are you? Why would I buy or trust equipment from - who are you again? How many years do you have in the industry? What warranty do I get and how can I be sure you (who are you) will back it?

I wouldnt worry too much about that. A project like this wouldnt aim to outsell Denon, it will be niche of course, but it could be a nice profitable niche. Again see KiSS' Divx players. Who where they? No one. But they ended up doing allright because they sold something customers (mostly geeks) wanted and no one else offered. Same for ouya. Oh and lets not forget Boxee! Nah, the problem will be designing and building the thing. If it works properly, it will sell.

[ixian]

If there's a word for "not even enough to be considered niche" that's what you could call this thing.

No company is going to build one. Hell the AVR market itself is dying out. Mod something yourself maybe, for the satisfaction? Though considering you can already get media players so small that they fit easily behind any TV (I have 3 myself) I don't really see the point to it. I mean, I have an AVR, and a tiny Zotac box behind my flat screen already, so technically, from a looks perspective....I already have this.

[Ned Scott]

This is basically what google was doing with the nexus Q. Too bad it seems that google has canned the product.

[Vertigo]

If there's a word for "not even enough to be considered niche" that's what you could call this thing.
No company is going to build one.

Based on XBMC? Perhaps not. With similar functionality? Im taking bets in 5 years you can buy an AVR with similar functionality from Pioneer, Phillips or the like. It still wont work nearly as well as XBMC though.

Hell the AVR market itself is dying out.

What makes you say that?
Last numbers I saw looked pretty healthy:
http://www.statista.com/statistics/23607...ed-states/

Almost $700M per year in the US alone, and growing.

BTW, I just stumbled upon this article:

Six reasons receivers shouldn't be media streamers
This year's AV receivers are poor choices as media streamers because of their mediocre graphical user interfaces, sluggish navigation, high cost of replacement, infrequent updates, and complex remotes.

http://news.cnet.com/8301-17938_105-2006...streamers/

The article is for the most part, making my point, except its drawing the wrong conclusion. Rather than saying receivers shouldnt be media streamers, they should say receivers have got to become better media streamers. "mediocre graphical user interfaces", "sluggish navigation", "infrequent updates", and "complex remotes", "No video streaming".

My points exactly. An embedded XBMC could take care of all that. And then some.

As for high cost of replacement, its a valid point, but an XBMC integrated AVR could be almost infinitely upgradable through software. You might even make it modular enough to allow upgrading hardware components, like the SoC board or the interface board.

Mod something yourself maybe, for the satisfaction?

I wish I could. I dont have the skills, but I'll see who and what I can find. If I can find some people with the right skill sets, I wouldnt mind investing some money to find out if its doable and work towards a prototype. But I would definitely approach it as a business opportunity, rather than to satisfy my own needs.

---

This is basically what google was doing with the nexus Q. Too bad it seems that google has canned the product.

AFAIK, NexusQ was just another settop box, like Boxee, no? Having something like that with XBMC might be a first step, but we pretty much have that already with Raspberry pie and the avalanche of android "stick PCs". I have yet to see something like that with a decent integrated amplifier or ability to connect and switch multiple HDMI sources though.

[Ned Scott]

The NexusQ was an Android powered settop box + amp for speakers. Someone even got XBMC running on it. It doesn't have any other AVR features, but considering GoogleTV boxes allow you to input another HDMI device into the box (basically allowing it to act as a switch between GoogleTV and a second HDMI device), it was conceivable that Google could add that to the NexusQ or a later version of it. So it would be a very simplified "AVR".

[Vertigo]

The NexusQ was an Android powered settop box + amp for speakers. Someone even got XBMC running on it. It doesn't have any other AVR features, but considering GoogleTV boxes allow you to input another HDMI device into the box (basically allowing it to act as a switch between GoogleTV and a second HDMI device), it was conceivable that Google could add that to the NexusQ or a later version of it. So it would be a very simplified "AVR".

Ah, interesting. I didnt realize google put an amp in there too! Wikipedia says you are right though, there was indeed a "stereo grade" stereo amp in there. Good thinking if you ask me .
Although for "my" box, with a focus on movies, at the very least I would provide a 3.1 setup and if doesnt add too much cost, preferably 5.1 or more.

[apertotes]

hello! i just registered on the site because of this thread. I am a complete noob on these things, so maybe what i am going to say is a stupidity. But isn't what the OP want something along Assassing Audiophile rigs?

http://assassinhtpc.com/audiophile.php

[Vertigo]

hello! i just registered on the site because of this thread. I am a complete noob on these things, so maybe what i am going to say is a stupidity. But isn't what the OP want something along Assassing Audiophile rigs?

As far as I can tell, those rigs are simply overpriced HTPCs with a lot of integrated "Bitstreaming lossless zero latency" bullshit. They dont have built-in amplifier.

This is closer to what I have in mind:
http://eumeles.wurmsdobler.org/systemoverview.html

But its only stereo, and aimed at music mostly.

[nickr]

So if I buy the right amp that does cec properly and doesn't drop hdmi audio, and I cable tie my acer revo on the back, I essntially have whay you seek.