[ilovemymac]

I am wondering if anyone who has used both can tell me how they compare. I know that roku has many channels that cost but I see they have some channels that play old tv content and I have trouble finding older stuff on Xbmc. I so realize in some ways they are like comparing apples and oranges both stream but still totally different.

I have been asked on Facebook when I talk about Xbmc by people who have a roku if they would prefer Xbmc and so forth. Since I have never owned one hoping people here can help.

[whitebelly]

They don't compare. Roku is a streaming device created for content owners to distribute via channels. XBMC is a media center for individuals to manage and play their own content, with addons that allow streaming and other "cool stuff". If you want Neftlix, Amazon, Hulu, etc, Roku works well. If you have a bunch of movie files, XBMC is wonderful.

[artrafael]

http://forum.xbmc.org/showthread.php?tid=156805

[rziulek]

I used XBMC until the Amazon Prime addon stopped working. As I hunted for another alternative, ROKU stood out as a favorite. So I bought ROKU as a Amazon Prime source only, but as I started using it I found there are more and more things that it did as well as or better than XBMC.

I installed PLEX media server with the ROKU 'channel' to handle all of my personal content and I have to admit that it is terrific - even if you want to use PLEX with XBMC as your front end for your media collection.

I then looked around for addons for ROKU that would give me other functionality that I had with XBMC. I found that like XBMC, ROKU has user created addons that are not supported by ROKU (ie: OneChannel, Ice Films, Free Cable, etc.). I also found a product that provides access to a lot of the websites of the broadcast and cable stations. It's called PlayOn. Unfortunately, there is a small fee associated with PlayOn but I found it worth every penny. They also have a companion product called PlayLater that gives you the capability of recording anything you can access through PlayOn.

The only thing I can't use ROKU for is live TV and my PVR setup. There is a ROKU front end for NextPVR, but it only provides playback, not the front end TV Guide or recording capabilities.

Overall, I love both of these alternatives. The tipping point for me was Amazon Prime. I watch Amazon Prime a lot so ROKU has an advantage currently. If the Amazon addon for XBMC is ever repaired, I will have to consider going back to XBMC.

[mkossler]

I've always rather thought of Roku and XBMC to be like an American muscle car compared to an Italian sports car. The muscle car is just as fast as the Italian, and handles a great variety of things the Italian does not - potholes, poor gas/petrol, etc. It may even be a bit more reliable. However, the is no matching the elegance of the Italian in terms of form, handling, and driving experience. One is utilitarian, the other passion. My .02, if you will - your mileage may vary.

Edit for bizarre double-entry of narrative text

[moejama]

I certainly would not consider XBMC more stable. It just more customizable, but it really misses on some major points, such as the library update engine. The fact XBMC still doesn't have a decent update engine is pretty sad. In contrast you can point Plex at your library and never have to worry about it again. The new Plex re-write is nice in that is separats the media server from the front-end.

I currently run both and while I prefer XBMC I also have more issues with it and it lacks the basic reliable functionality of Plex, with some tweaking you can get MOST of that but sadly you are relying add-ons that break or sometime have memory leaks and all that fun stuff.

Plex, as a basic media server, is superior. It scrapes media better out of the box, the web interface is much easier to use than setting media sources from within the XBMC and the library update of Plex is far superior to that of XBMC. Unfortunately Plex is still fresh out of it's re-write so it has few options beyond the basics. The problem I see with Plex is they have no focus or motivation toward the PC front-end piece of the software, the Plex Media Center or Plex Player.

Their business model is all about getting people to buy the Plex app for mobile devices and set-top boxes. I prefer the XBMC model, but the simple fact is the XBMC development team is too spending a lot of time working on advanced features when they should be prioritizing basic functionality. Things like a central media database should not require a MySQL setup, the XBMC updater should be able to update content as it changes and only add the new content, not have to re-scan the entire folder. Simple features like play next file in a series should be tested so they PLAY not just select the file. Currently and with the latest Gotham release if you have it set to load the show information on selection the play next video doesn't work. That feature has been broken for years. Having to do in an edit my advancedsetting.xml to fix their new date system is something they should add to the interface.

I think Plex is going the right direction with the re-write and the web interface. This is the direction most quality software is going. XBMC has too much legacy code in it and the bugs take too long to fix. Sadly Plex has way less support, way less documentation and it's still in beta since the re-write, still it works great when combined with something like Roku or CoStar.

XBMC is way prettier and customizable, but when you start adding the add-ons it can slow way down and leak memory very easily. Using watchdog for instance last night it was using 3.4 gigs of memory, all because the dev team is busy focusing on 3D video code instead of getting the core code perfected. A central database option and a smart update engine are necessary for XBMC to remain relevant in a world where media boxes are becoming common and people expect them to just work. Better built in scrapping would be ideal, a more user friend means to add libraries would help reduce a lot of the spam we see on questions people ask. The fact that plex doesn't require all this SMB share stuff is great for most people and it completely bypasses all the issues with people not know how to network and passwords and homegroups getting in the way with every new release of windows. That should be taken into account since the vast majority of XBMC installs are on Windows and most media server are on a Windows OS. Plex bypasses all that BS by using UPNP which we can see is the direction media players are going.

I think XBMC needs to separate the server from the front-end just like Plex did and provide a UPNP just like Plex did. No more need to run a separate SQL database, no complicated SMB shares and so on and so forth. That's just the direction the world is going. If plex is smart they will soon add in most of the lacking features that will make their platform customizable again and the ease of use will draw in developers. I'd rather see XBMC win this battle, but the development team is just making the wrong decisions as far as creating a solid media experience vs trying to add every bell and whistle to some old ass legacy code. 3D movies are just not that popular, but look at how much time people spending fighting with the XBMC library. All their efforts should first go to making the library function seamlessly and update on it's own with the least performance impact. That's the core of any good media system and users are spending too much time having to work around the built in limitations via external programs and half working add-ons.

I would certainly pay money for every a yearly fee for a more robust version of XBMC with the Plex style library update and superior scrapping. The transcoding is a nice touch, but for me unnecessary. The ability to link the library through UPNP instead of MySQL is awesome. Plex isn't even a ram hog from what I can tell, i've certainly never caught it using 3.4 gigs of memory just because I installed one add-on. All this needs to be added right into the basic install of XBMC instead of expecting people to spend days figuring out what they need to get in order to automate and scrape a library and thus have a working media center.

On a side not, American muscle cars are hardly something I'd call reliable. I have one, they are not reliable. My Honda is reliable because it only does what I need it to do, which is drive me places. In the end that's all I need a media center to do, have a share library that scrapes and updates. It's not asking much for the dev team to focus on that first and foremost. The only other feature I need are skins, all the video plug-ins are primarily crap. I just use sickbeard, couchpotato and Sabnzbd to do all the rest of the work and that setup too was a bitch, but it's solid once you get it all running. A game launcher is a nice touch, but that can all be handled externally and is probably more reliable that way.

That also brings up a good example of why you need to separate the front end and back end. Some day XBMC will support UPNP fully, but nobody will want to be required to run the XBMC front-end merely to provide sharing to their other XBMC clients. If I want to play a game on my media center I may want to close down XBMC and unlike Plex if I close the front-end I close everything including the UPNP server and all the automatic scrapping. I feel more at home with XBMC, but the Plex software runs better and runs smarter and the setup couldn't be much easier especially considering it's only beta. The biggest complaint I have is no gamepad support. I have to use a third party program to map the gamepad to keyboard keys, but again that's where you see Plex is focuses more on supporting Android devices and Roku boxes and such where you have a remote. I suspect all these small things will be fixed as it comes out of beta and I'ld be happy to pay for the program to help ensure a healthy development cycle.

Now if we could only get Hulu, Netflix and Amazon Prime, but that's where the Roku + Plex combo cannot be beat. If you want any of those three services the Roku is, by far, your best bet for the money and for a fast and reliable interface. Perhaps it was smart of them to let Plex do all the library services in order to keep their support needs minimal because scrappers, library updates and networking questions are abundant with media libraries. I certainly don't mind pay the 7 dollar or whatever for the Plex app, even if the menu is very plain and basic compared to the XBMC.

I will continue to use XBMC since my TV has Hulu and Amazon built in even though my need for Hulu is mostly gone, but for any new room setups I'll be going with Roku and Plex simply to make my life easy and to save me from troubleshooting. I'm hoping to see Google TV mature into something more reliable as I like the idea of having access to Android apps, but for now it just looks like bugs and slow performance. I also have no need of services like Chromecast since the browser interface is not designed for a remote.

[Ned Scott]

@moejama

There's a somewhat common misconception about how XBMC developers work. People see work being done on one feature and assume that this takes resources away from other features or bug fixes. This simply isn't true.

For example, I think 3D is pretty dumb in most movies, and I would not consider buying a 3D TV or paying such features any mind. However if someone else likes 3D and wants to work on that area, that is no skin off my back (I'm not a developer myself, but the example is still the same). The team as a whole is not focusing on this or that feature, and such features are only worked on if someone wants to work on it.

If the watchdog add-on has an issue where it is using 3 GB of ram then you need to talk to the watch dog add-on author, not Team XBMC. If it's the result of a python issue then even that would have nothing to do with the devs who are working on 3D. One or two devs who occasionally work on that one feature does not equal the whole team.

Regarding library sharing, even as-is, Gotham has an excellent UPnP server built-in that replaces most of the functionality of MySQL. In the future even more improvements will be coming on that front. I don't even touch MySQL anymore.

I know this doesn't address everything you've said here, but the point is that there's a lot of misunderstanding on your part about why things are the way they are. They are still that way, so I'm not disputing Plex's ease of set up or any of XBMC's shortcomings.

XBMC is a fine balance between powerful customization options and usability. I think you'll be pleasantly surprised at what XBMC has in store for those usability options, and how they might be solutions you didn't expect.

[mkossler]

I certainly would not consider XBMC more stable... <snip!>

On a side not, American muscle cars are hardly something I'd call reliable. I have one, they are not reliable. My Honda is reliable because it only does what I need it to do, which is drive me places. In the end that's all I need a media center to do, have a share library that scrapes and updates. It's not asking much for the dev team to focus on that first and foremost. The only other feature I need are skins, all the video plug-ins are primarily crap. I just use sickbeard, couchpotato and Sabnzbd to do all the rest of the work and that setup too was a bitch, but it's solid once you get it all running. A game launcher is a nice touch, but that can all be handled externally and is probably more reliable that way.

<snip!> The biggest complaint I have is no gamepad support. I have to use a third party program to map the gamepad to keyboard keys, but again that's where you see Plex is focuses more on supporting Android devices and Roku boxes and such where you have a remote. I suspect all these small things will be fixed as it comes out of beta and I'ld be happy to pay for the program to help ensure a healthy development cycle.

Now if we could only get Hulu, Netflix and Amazon Prime, but that's where the Roku + Plex combo cannot be beat. If you want any of those three services the Roku is, by far, your best bet for the money and for a fast and reliable interface. Perhaps it was smart of them to let Plex do all the library services in order to keep their support needs minimal because scrappers, library updates and networking questions are abundant with media libraries. I certainly don't mind pay the 7 dollar or whatever for the Plex app, even if the menu is very plain and basic compared to the XBMC.

I will continue to use XBMC since my TV has Hulu and Amazon built in even though my need for Hulu is mostly gone, but for any new room setups I'll be going with Roku and Plex simply to make my life easy and to save me from troubleshooting. I'm hoping to see Google TV mature into something more reliable as I like the idea of having access to Android apps, but for now it just looks like bugs and slow performance. I also have no need of services like Chromecast since the browser interface is not designed for a remote.

Heck of a Plex vs XBMC narrative for a Roku vs XBMC thread

On a side note, your argument - "I have an American car, and they're not reliable" is a classic example of debate by anecdote. While your experience with you car may have been less than stellar (and your opinion valid in that context), it is but one data point among millions, and does not, in and of itself, make your opinion into fact. I can assure you, however, that the data shows American muscle cars in general to be far more reliable - comparatively - than Italian sports cars.

Regardless, I enjoyed reading your post.

[shortskoolbus]

I certainly would not consider XBMC more stable. It just more customizable, but it really misses on some major points, such as the library update engine. The fact XBMC still doesn't have a decent update engine is pretty sad. In contrast you can point Plex at your library and never have to worry about it again. The new Plex re-write is nice in that is separats the media server from the front-end.

I currently run both and while I prefer XBMC I also have more issues with it and it lacks the basic reliable functionality of Plex, with some tweaking you can get MOST of that but sadly you are relying add-ons that break or sometime have memory leaks and all that fun stuff.

Plex, as a basic media server, is superior. It scrapes media better out of the box, the web interface is much easier to use than setting media sources from within the XBMC and the library update of Plex is far superior to that of XBMC. Unfortunately Plex is still fresh out of it's re-write so it has few options beyond the basics. The problem I see with Plex is they have no focus or motivation toward the PC front-end piece of the software, the Plex Media Center or Plex Player.

Their business model is all about getting people to buy the Plex app for mobile devices and set-top boxes. I prefer the XBMC model, but the simple fact is the XBMC development team is too spending a lot of time working on advanced features when they should be prioritizing basic functionality. Things like a central media database should not require a MySQL setup, the XBMC updater should be able to update content as it changes and only add the new content, not have to re-scan the entire folder. Simple features like play next file in a series should be tested so they PLAY not just select the file. Currently and with the latest Gotham release if you have it set to load the show information on selection the play next video doesn't work. That feature has been broken for years. Having to do in an edit my advancedsetting.xml to fix their new date system is something they should add to the interface.

I think Plex is going the right direction with the re-write and the web interface. This is the direction most quality software is going. XBMC has too much legacy code in it and the bugs take too long to fix. Sadly Plex has way less support, way less documentation and it's still in beta since the re-write, still it works great when combined with something like Roku or CoStar.

XBMC is way prettier and customizable, but when you start adding the add-ons it can slow way down and leak memory very easily. Using watchdog for instance last night it was using 3.4 gigs of memory, all because the dev team is busy focusing on 3D video code instead of getting the core code perfected. A central database option and a smart update engine are necessary for XBMC to remain relevant in a world where media boxes are becoming common and people expect them to just work. Better built in scrapping would be ideal, a more user friend means to add libraries would help reduce a lot of the spam we see on questions people ask. The fact that plex doesn't require all this SMB share stuff is great for most people and it completely bypasses all the issues with people not know how to network and passwords and homegroups getting in the way with every new release of windows. That should be taken into account since the vast majority of XBMC installs are on Windows and most media server are on a Windows OS. Plex bypasses all that BS by using UPNP which we can see is the direction media players are going.

I think XBMC needs to separate the server from the front-end just like Plex did and provide a UPNP just like Plex did. No more need to run a separate SQL database, no complicated SMB shares and so on and so forth. That's just the direction the world is going. If plex is smart they will soon add in most of the lacking features that will make their platform customizable again and the ease of use will draw in developers. I'd rather see XBMC win this battle, but the development team is just making the wrong decisions as far as creating a solid media experience vs trying to add every bell and whistle to some old ass legacy code. 3D movies are just not that popular, but look at how much time people spending fighting with the XBMC library. All their efforts should first go to making the library function seamlessly and update on it's own with the least performance impact. That's the core of any good media system and users are spending too much time having to work around the built in limitations via external programs and half working add-ons.

I would certainly pay money for every a yearly fee for a more robust version of XBMC with the Plex style library update and superior scrapping. The transcoding is a nice touch, but for me unnecessary. The ability to link the library through UPNP instead of MySQL is awesome. Plex isn't even a ram hog from what I can tell, i've certainly never caught it using 3.4 gigs of memory just because I installed one add-on. All this needs to be added right into the basic install of XBMC instead of expecting people to spend days figuring out what they need to get in order to automate and scrape a library and thus have a working media center.

On a side not, American muscle cars are hardly something I'd call reliable. I have one, they are not reliable. My Honda is reliable because it only does what I need it to do, which is drive me places. In the end that's all I need a media center to do, have a share library that scrapes and updates. It's not asking much for the dev team to focus on that first and foremost. The only other feature I need are skins, all the video plug-ins are primarily crap. I just use sickbeard, couchpotato and Sabnzbd to do all the rest of the work and that setup too was a bitch, but it's solid once you get it all running. A game launcher is a nice touch, but that can all be handled externally and is probably more reliable that way.

That also brings up a good example of why you need to separate the front end and back end. Some day XBMC will support UPNP fully, but nobody will want to be required to run the XBMC front-end merely to provide sharing to their other XBMC clients. If I want to play a game on my media center I may want to close down XBMC and unlike Plex if I close the front-end I close everything including the UPNP server and all the automatic scrapping. I feel more at home with XBMC, but the Plex software runs better and runs smarter and the setup couldn't be much easier especially considering it's only beta. The biggest complaint I have is no gamepad support. I have to use a third party program to map the gamepad to keyboard keys, but again that's where you see Plex is focuses more on supporting Android devices and Roku boxes and such where you have a remote. I suspect all these small things will be fixed as it comes out of beta and I'ld be happy to pay for the program to help ensure a healthy development cycle.

Now if we could only get Hulu, Netflix and Amazon Prime, but that's where the Roku + Plex combo cannot be beat. If you want any of those three services the Roku is, by far, your best bet for the money and for a fast and reliable interface. Perhaps it was smart of them to let Plex do all the library services in order to keep their support needs minimal because scrappers, library updates and networking questions are abundant with media libraries. I certainly don't mind pay the 7 dollar or whatever for the Plex app, even if the menu is very plain and basic compared to the XBMC.

I will continue to use XBMC since my TV has Hulu and Amazon built in even though my need for Hulu is mostly gone, but for any new room setups I'll be going with Roku and Plex simply to make my life easy and to save me from troubleshooting. I'm hoping to see Google TV mature into something more reliable as I like the idea of having access to Android apps, but for now it just looks like bugs and slow performance. I also have no need of services like Chromecast since the browser interface is not designed for a remote.

That's a nice writeup there. I love the interface of XBMC, some of the skins these guys make, and that's why I use it. The fanart back grounds, the list views, etc. But I wish the experience could be more seamless which you've more or less said. I never really considered PLEX, but i do see a lot of people like it as an alternative. Not sure really about the transcoding thing, don't know much about this but I can play everything fine with my setup as it is on XBMC. I don't really need to be able to watch movies on my 5" android phone either, so not a big deal there. I guess i need an always on machine for Plex server too, which the only always on machines i have are my readynas and rpi, and neither will handle transcoding very well.

I've looked at the Roku boxes, but they just can't match the awesomeness of having dynamic fan art backgrounds as you scroll through movies. It seems like a good alternative if you just need something that works, and I don't really mind paying a small subscription to Hulu or whatever it is. But really I think the whole experience is just kind of dumbed down.

The part i wonder about is, if Roku can get Hulu, Amazon, etc on board, why can't XBMC? What does it take?

[whitebelly]

The part i wonder about is, if Roku can get Hulu, Amazon, etc on board, why can't XBMC? What does it take?

The perception of safety through DRM on a closed platform for starters.

[shortskoolbus]

The part i wonder about is, if Roku can get Hulu, Amazon, etc on board, why can't XBMC? What does it take?

The perception of safety through DRM on a closed platform for starters.

'
I mean, I guess that Roku clearly licenses some kind of DRM technology which is deemed acceptable to the content providers, which in turn they allow them to access their content from the box. I guess what I'm saying is what would it take for XBMC to "officially" get Hulu, Netflix, etc up and running? If Roku, boxee, playstation, xbox, etc these guys can do it, why can't XBMC? I have no idea what it takes.. just wondering out loud.

[artrafael]

If Roku, boxee, playstation, xbox, etc these guys can do it, why can't XBMC? I have no idea what it takes..

Selling our soul to the devil?

[shortskoolbus]

If Roku, boxee, playstation, xbox, etc these guys can do it, why can't XBMC? I have no idea what it takes..

Selling our soul to the devil?

Haha. Sometimes I just want it to work. More and more Usenet is returning bad downloads with the takedowns, certain video plugins while I appreciate them break a lot of times, can't guarantee the source will work or give me the quality I prefer.... Just saying it would be nice to have those options, in addition to all the great things xbmc does now. Guess it would be like an "official" Amazon prime plugin, for example.

[artrafael]

XBMC is open source, so no DRM. Without the latter, the other boys won't let us play ball with them.

[Ned Scott]

The part i wonder about is, if Roku can get Hulu, Amazon, etc on board, why can't XBMC? What does it take?

Money.

If someone could convince them that the risk was worth the reward, they would make it happen. The industry doesn't believe they can do it without DRM. Which is funny considering you could get Amazon via an unofficial add-on that only worked for paid subscribers, and they had no real evidence that they were losing money or being harmed, but they still let their fears lead them to enabling DRM.

They also fear not having total control over the user interface. XBMC allows a custom interface. Any dealings with Hulu, Amazon, Netflix, etc would involve some control or assurance regarding a consistent user interface for their services.