HTPC Suggestions and Discussions

[thideras]

Hello all,

First, I would like to apologize if this gets to be lengthy. I seem to be good at writing walls of text for simple ideas. I talked this over with IMOG before making this post, didn't want to be out of place in your guys "haven".

I was looking over one of the recent posts on our front page since I'm starting to be interested in the HTPC market. I'm fairly new to this section of computing and have done a little bit of research already. There is a lot to take in and most reviews of hardware don't seem to cover everything. For example, I'll use the recent review of the ATI HD 5550. I'm not picking on the author (<3 you Splat), but it is a good example. I started reading the article and it came across as "This is an HTPC card and I will review it as such."

This should add a small boost in performance over the GDDR3 and DDR2 models while retaining the low-power profile, perfect for inclusion in a Home Theater PC (HTPC) build.
The Radeon HD 5550 is an entry level card aimed at the HTPC market. The most notable features are it’s very low power draw (39W maximum) and hardware acceleration for video. If you throw a Bluray Disc Drive in with it, then you can also stream the uncompressed audio directly to your receiver without the need for another audio card. All in all, the specs add up to a decent option for a HTPC and possibly even worthy of a low end gaming computer with it’s DirectX 11 support. When you consider this card has a MSRP of $75, it begins to look even better.The tests and results were a bit confusing to me. The 3d tests are good, no complaints there. The only "video" ones tested were Youtube and Hulu. From my understanding, Flash is not accelerated by the video card and your tests basically prove that (CPU usage tests). For the review being so HTPC directed, it doesn't have many HTPC tests along with the results are muddled. From what the article says, I wouldn't benefit from running that card and in some cases, it slightly increases the CPU load. I would like to see the difference in programs such as Windows Media Center, MythTV, Media Portal, XMBC, Boxee, VLC and Media Player Classic. From my understanding, these are the programs that people use on their HTPC's and would make the review a lot more solid. Personally, I use Media Portal, Boxee and VLC to view videos. I would be extremely interested in a video card that can make it look better.

The problem is, I don't have all these answers. I don't have experience testing these different software programs to see if there was an improvement. I'm sure that if we put our heads together, we can hammer something out and get quite a few articles from it. For example, I have yet to see a length, well written article that explains the pros and cons between all those programs then actually uses them for longer than 5 minutes. It could be staged over a few weeks (or months) with a large write-up at the end explaining everything.

------------------

Ok, on to ideas. I've given a few examples for programs to use for video:



MythTV
Windows Media Center
Media Portal
XMBC
Boxee
VLC
Media Player Classic


You could split these up by operating system (Windows vs Linux) or bundle them all together. You could do remote servers with MythTV to see if it improves speed for slower systems. You could compare the Linux and Windows versions of XMBC to see if there is a difference. There are a lot of combinations that you could do that few (or no) websites cover. I know that I couldn't find a lot of information that I was looking for and basically had to "just try it".

So, how do you test these programs? Back to Splat's article, I think that CPU usage is a very good indicator on how much load is being put on the video card and how much it is "saving" on the CPU. We may be able to add a program to record FPS (such as Fraps) to make sure that the content is smooth (when V-Sync is on) or to see how "high" the FPS goes. Is there a program that can quickly (also with low load) record CPU usage to a log file automatically? Similar to what CoreTemp does.

I may be able to take some time and get things working, but I need a computer that has the ability to actually play videos better than my Dell Dimension that I got for free (2.4ghz single core P4, integrated video). I hope you guys are able to take these suggestions and run with them as it should give you a few "free" article ideas and more page views. I'm fully willing to sit down and discuss any ideas you have to bring to the table and sacrifice a computer or two for testing purposes.

[thideras]

I'm surprised at the lack of responses.

[I.M.O.G.]

Last week was not a good week time-wise for the news team - none of us who are typically active every day had any time. :-/

I'll be getting back to this however.

[thideras]

I was beginning to question if I put it in the folding forum or something.

EDIT: I could see this being taken incorrectly. I meant it as "I put it in the wrong forum and people avoided the thread since I'm stupid".

[thideras]

:sn:

[I.M.O.G.]

Anyone else with input? Haven't had time to get to this myself yet...

[Tsnowflake]

An option some people do is to forgo the internal player in those programs and use an external one like VLC or MPC-HC. I know media portal allows you to set the player it uses, and I know that MPC-HC has a whole bunch of filters, shaders. sharpness settings, and color adjustments you can play with.

MPC-HC also has a built in FRAPS like feature that can tell you if you are skipping/losing frames.

As for testing you can do some basic contrast/sharpness tests with AVSHD709 (http://www.avsforum.com/avs-vb/showthread.php?t=948496)
There is a color test in there but from what I understood you needed a set of filters which I didn't have. (Physical light filters not a computer generated filter to avoid any potential confusion lol)

Cheese Slices (http://www.avsforum.com/avs-vb/showthread.php?t=1157287) can be used to test various types of de-interlacing.

hope this helps!

[jhanby]

First off, thanks for the feedback !

I think the main hurdle with the testing which you are proposing is time. I for one have noticed a massive lull in frontpage content recently, but this is a busy time of the year with holidays and covering other peoples holidays. (EDIT : this wasn't meant in a bad way, I know that I haven't been giving this much attention either due to lots of work recently).

So the first issue is going to be spending the amount of time on a single, massive article testing each "Media Centre" with each player on each OS. This also means that you can't use it as normal day to day HTPC, because you have to empirically test it when you change it.

The second issue I can forsee is that if one article was created testing a single graphics cards with all the possible combinations of Media Centres, Players and OSs. It is going to be a huge article, or a not very in depth article. There are two solutions to this. The first is to split the article into pages, one for each combination. This would make the article less daunting than a single continuous page of writing. It would also let people who want to know about one specific "combination" navigate to said combination. The problem with pages is that you can't just have one article in pages and the rest as a single page, you'd have to change them all which I know people aren't keen on.

The second solution would be to split it into multiple articles. Maybe one dedicated to all things unix, one dedicated to a each Windows Media centre. The downside to this is that you can't really compare them, because you keep having to refer back to other articles. However, this does mean that other people can help and split the workload around.

Don't get me wrong, I'm all up for this idea and will help as much as I can. For it to work though it's going to have to be very thoroughly organised so that everybody is singing from the same song sheet.

If it does go along, I'm sure we can ask the other prolific HTPCers (Land Shark, Wiked Clown, etc etc) to help out and brainstorm ideas.

[thideras]

Were any articles created from this?

[jhanby]

Were any articles created from this?

I don't think so, but as there is a massive lack of articles from the HTPC section, we should put some effort into getting this going (maybe I haven't been pushing enough in the section :shrug: )

I'd be happy to contribute to this as well. I have an alright HTPC card I think and would happily run tests on it. Don't think I'm exprienced enough to write a whole article just yet on it.

I'll PM some HTPC guys pointing at this thread and ask that they give their thoughts on it.

We'd definitly need some sort of set article goals, as I'd hate to see one huge article started on with no real aim.

Maybe start with the tittle of the article and write a rough framework ?

[I.M.O.G.]

Would love to see this developed into front page articles. :)

[splat]

FYI, yes, flash 10 does have hardware acceleration. As does video encoding/decoding in some players. That's why those were tested in that article. The error in my original tests was that ATI did not make it clear that to enable the hardware acceleration, you have to install a 2nd package that is separate from the regular CCC driver package.

ninja edit: and for HTPC purposes, thats really all you can test a GPU for - cpu usage while decoding/encoding. if you can think of other tests by all means please suggest them.