AVS › AVS Forum › Video Components › Home Theater Computers › Sandy Bridge vs. MPC-HC
New Posts  All Forums:Forum Nav:

Sandy Bridge vs. MPC-HC

post #1 of 22
Thread Starter 
Is there any word on when a fix (if ever) will be released for the two to work nicely together. I'm currently using ffdshow-tryouts, but would like to go back to DXVA if possible. Using anything other media player isn't an option, mpc-hc just works so well and cleanly.
post #2 of 22
http://www.xvidvideo.ru/media-player...-5-2-3085.html

You can try the latest MPC-HC build and see how it goes.
post #3 of 22
I also had problems with MPC HC stable version, download the latest beta of the program and DXVA operate normally.

I have a i5 2500 on a Asus P8H67 M EVO

But despite getting DXVA, the main problem is the poor quality of viewing on the MPC, too many macroblocks at the bottom of the image, unacceptable to some mkv (4 674Kbps-720P), another videos it plays relatively well.

I come from itself uses a HD4550 and initially I felt somewhat frustrated.

These days I'm testing VLC and MPC videos that they looked bad in VLC can see a great improvement, not macroblocks noteworthy.
VLC has never been among my favorite players, but until the HD2000 INTEL and MPC HC does not get along, should be unfaithful.

Have you tried other video players?

Sorry for my googlenglish
post #4 of 22
Excuse my ignorance but what is the advantage to DXVA when you have such a capable processor? Isn't hardware acceleration intended for slower processors? There must be more to it than just that...?
post #5 of 22
You'd better use (for now)

- Microsoft DTV-DVD Video Decoder for H.264 and MPEG-2 (DXVA is supported)
- WMVideo Decoder DMO for VC-1 (DXVA is not supported, but post-processing, including hardware deinterlacing, works)

Both are built in Windows.

DXVA includes not only hardware decode acceleration, but also hardware post-processing, in particular interlacing. Decoding is easy for any decent dual-core CPU. But there is no satisfactory software deinterlacing solution. For example try to play VC-1 1080i Cheese Slices with ffdshow Video Decoder (wmv9) and with WMVideo Decoder DMO and compare them. You will see the latter is a lot better.
post #6 of 22
Quote:
Originally Posted by GatoEnFuego View Post

Excuse my ignorance but what is the advantage to DXVA when you have such a capable processor? Isn't hardware acceleration intended for slower processors? There must be more to it than just that...?

If you use hardware acceleration on the i3, the CPU will basically step down to idle allowing you to build a machine that can watch blu-rays at about 15-19 watts.. not much more than your average STB. If you don't use hardware acceleration then you will consume 45+ watts and probably have more fan noise and other crap like that.
post #7 of 22
BTW actual measurements here (by Kill-A-Watt):

- System: Core i3-2100, GA-H67MA-UD2H, 2 x 2GB DDR3-1333, 1 x HDD, MODU82+

- Idle: 38W
- H.264 playback (DXVA): 43W
- H.264 playback (software): 48W
post #8 of 22
Quote:
Originally Posted by renethx View Post

BTW actual measurements here (by Kill-A-Watt):

- System: Core i3-2100, GA-H67MA-UD2H, 2 x 2GB DDR3-1333, 1 x HDD, MODU82+

- Idle: 38W
- H.264 playback (DXVA): 43W
- H.264 playback (software): 48W

Thats not bad for the standard i3-2100, my kill-o-watt averages were from the i3-2100t version
post #9 of 22
Quote:
Originally Posted by byronmhome View Post

Thats not bad for the standard i3-2100, my kill-o-watt averages were from the i3-2100t version

You must be using a very good PSU (i.e. very efficient at low output). PicoPSU?
post #10 of 22
Quote:
Originally Posted by renethx View Post

You must be using a very good PSU (i.e. very efficient). PicoPSU?

Nope, just one of those earth watt 380 psus. Buying a pico-psu would eat up more $$ than the energy savings their worth. Maybe if i was off grid i would get one of them

TDP of the i3-2100t is 30 watts less than i3-2100. Are you running the t variant as well? your numbers jive with the non t version.
post #11 of 22
I tested 2100T too and never got such a good result (only a few watts lower than 2100 at idle, 10-15W lower at load). This agrees with the results by several 2100T users. On the other hand, some users reported much better results (like you).

Here are actual measurements of the power consumption of a 2100T system at idle by Japanese users (the AC voltage is 100V so that the efficiency of PSU is worse than 120V/225V).

- ASRock H67M-ITX, 2 x 2GB, 1 x 3.5" HDD, pico-PSU150T: 26W
- BIOSTAR H61MH, 2 x 2GB, 300W 80 PLUS GOLD: 23W
- Foxconn H67S-B3, 2 x 2GB, 1 x 2.5" HDD, pico-PSU120: 20W
- ASRock H67M-ITX, 2 x 2GB, 1 x SSD, IN WIN IP-AD160-2 160W: 28W
- Jetway MIH67M-HU3, 2 x 2GB, 1 x SSD, TX850W: 36W
- ASRock H67DE3 2 x 1GB, 1 x SSD, PSU?: 46W
- ASUS P8H67-M EVO, 2 x 2GB 1.35V, 1 x 3.5" HDD, 750W 80 PLUS Silver: 39W

So the result depends on lots of factors (motherboard, HDD, PSU, and maybe the chip itself). On the other hand the power consumption at load is uniformly much higher, close to that of 2100 (one shouldn't expect 65W-35W = 30W difference).
post #12 of 22
Quote:
Originally Posted by renethx View Post

I tested 2100T too and never got such a good result (only a few watts lower than 2100 at idle, 10-15W lower at load). This agrees with the results by several 2100T users. On the other hand, some users reported much better results (like you).

Here are actual measurements of the power consumption of a 2100T system at idle by Japanese users (the AC voltage is 100V so that the efficiency of PSU is worse than 120V/225V).

- ASRock H67M-ITX, 2 x 2GB, pico-PSU150T, 1 x 3.5" HDD: 26W
- BIOSTAR H61MH, 2 x 2GB, 300W 80 PLUS GOLD: 23W
- Foxconn H67S-B3, 2 x 2GB, 1 x 2.5" HDD, pico-PSU120: 20W
- ASRock H67M-ITX, 2 x 2GB, 1 x SSD, IN WIN IP-AD160-2 160W: 28W
- Jetway MIH67M-HU3, 2 x 2GB, 1 x SSD, TX850W: 36W
- ASRock H67DE3 2 x 1GB, 1 x SSD, PSU?: 46W
- ASUS P8H67-M EVO, 2 x 2GB 1.35V, 1 x 3.5" HDD, 750W 80 PLUS Silver: 39W

So the result depends on lots of factors (motherboard, HDD, PSU, and maybe the chip itself).

Perhaps he was using an Intel motherboard? Para-phrasing one of SPCR's reviews, Intel boards are partially powered by fairy dust.
post #13 of 22
I use Win7 x32, shark007 and arcsoft are the only software installed. Nothing running in taskbar, no notification icons/programs or any of that. I just have a single 2tb wd green drive as well.. I use AS Rock h67m-itx.. its rev3, same as intel mitx for the most part. Computer was the only thing plugged into my killowatt. There are also no running case fans on my case, just the PSU and CPU cooler.
post #14 of 22
Quote:
Originally Posted by ilovejedd View Post

Perhaps he was using an Intel motherboard? Para-phrasing one of SPCR's reviews, Intel boards are partially powered by fairy dust.

hahaha.. i didn't want to even build an Intel system so i felt bad just going with an intel cpu I wanted my amd zacate build to work, but it will have to wait until SL5 gets ironed out more for gpu acceleration.. just not building out a low power htpc for 400+ that can't even play netfilx or wife would kill me
post #15 of 22
Quote:
Originally Posted by renethx View Post

- ASRock H67M-ITX, 2 x 2GB, 1 x 3.5" HDD, pico-PSU150T: 26W
- BIOSTAR H61MH, 2 x 2GB, 300W 80 PLUS GOLD: 23W
- Foxconn H67S-B3, 2 x 2GB, 1 x 2.5" HDD, pico-PSU120: 20W
- ASRock H67M-ITX, 2 x 2GB, 1 x SSD, IN WIN IP-AD160-2 160W: 28W
- Jetway MIH67M-HU3, 2 x 2GB, 1 x SSD, TX850W: 36W
- ASRock H67DE3 2 x 1GB, 1 x SSD, PSU?: 46W
- ASUS P8H67-M EVO, 2 x 2GB 1.35V, 1 x 3.5" HDD, 750W 80 PLUS Silver: 39W

So the result depends on lots of factors (motherboard, HDD, PSU, and maybe the chip itself). On the other hand the power consumption at load is uniformly much higher, close to that of 2100 (one shouldn't expect 65W-35W = 30W difference).

I think the pico psu is the highest source responsible as renethx said in a post above, the efficiency curve at low power is critical to saving energy.

In my case with a i5 2500 + ASUS P8H67-M EVO + 2x4 Gskill + 1 SSD + Seasonic X560 +OD: I managed to see an idle consumption from 29W to 36W, consumption varied too much to categorically state the real value.

Carried out, more tests of different types to post a blog entry about the use and behavior of this type of platform.
post #16 of 22
Quote:
Originally Posted by renethx View Post

You'd better use (for now)

- Microsoft DTV-DVD Video Decoder for H.264 and MPEG-2 (DXVA is supported)
- WMVideo Decoder DMO for VC-1 (DXVA is not supported, but post-processing, including hardware deinterlacing, works)

Both are built in Windows.

DXVA includes not only hardware decode acceleration, but also hardware post-processing, in particular interlacing. Decoding is easy for any decent dual-core CPU. But there is no satisfactory software deinterlacing solution. For example try to play VC-1 1080i Cheese Slices with ffdshow Video Decoder (wmv9) and with WMVideo Decoder DMO and compare them. You will see the latter is a lot better.

Adding another option:

If you have a copy of TMT around you can do this to get DXVA support in MPC-HC in Intel GPUs.
post #17 of 22
Quote:
Originally Posted by jmqnick View Post

I think the pico psu is the highest source responsible as renethx said in a post above, the efficiency curve at low power is critical to saving energy.

In my case with a i5 2500 + ASUS P8H67-M EVO + 2x4 Gskill + 1 SSD + Seasonic X560 +OD: I managed to see an idle consumption from 29W to 36W, consumption varied too much to categorically state the real value.

Carried out, more tests of different types to post a blog entry about the use and behavior of this type of platform.

With an i5-2500 idle load is usually 39-40watt on a good build.. 29/36w would be a great idle load to brag about for that box.

The only machine i'm tempted to build in comparison is a sempron 140 on 880g motherboard. They're so cheap and i've seen a few people build WHS with them at 31-37watt idle with junk parts.
post #18 of 22
Thread Starter 
So far the latest beta version 1.5.2.3085 is working quite nicely.
post #19 of 22
Well, the latest beta version of MPC DXVA works, but image quality leaves much to be desired, at least in my case.

I have installed Windows 7 without any codec except for the system, installed MPC HC and still is horrible in almost all cases, the definition is very bad.

What is your combination of video codecs?

I have also tried with VLC, without DXVA is perfect, but when set on for hardware acceleration start macroblocks.
The graphics drivers are updated to the latest version
post #20 of 22
Quote:
Originally Posted by renethx View Post

BTW actual measurements here (by Kill-A-Watt):

- System: Core i3-2100, GA-H67MA-UD2H, 2 x 2GB DDR3-1333, 1 x HDD, MODU82+

- Idle: 38W
- H.264 playback (DXVA): 43W
- H.264 playback (software): 48W

Measured with digital voltmeter and ammeter.

- System: i7-2600k (stock CPU cooler), Intel DH67GD, 2 x 4GB DDR3-1333 (Samsung MV-3V2G4), 2 x HDD (WD500BEVT and WD20EVDS), Liteon Blu-ray burner (iHBS212), 2 x 120mm case fans set on low (Stock Antec NSK-2480), PSU Seasonic X-660.

- Idle 38 to 40W
- Blu-ray ISO playback with IGP 40 to 45 watts.
LL
post #21 of 22
I have a i3-2100, asrock h67m-itx, 2x2gb ddr3 dimms, crucial c4 64gb ssd. antec isk-310 150 case. idle on the windows desktop it uses 36-37watts and playing a 720p mkv it uses 40 watts in xbmc.
the newest svn from xvidvideo.ru of mpc hc works with dvxa but the newest xbmc nightly doesn't. an openelec nightly worked with vaapi but had some weird visual artifact problems that only happened on some files. 720p mkv playback under openelec used less power.. about 37 watts.
edit: measured with a kill a watt
post #22 of 22
Alright, i put together an MSI 880gn-e35 + AMD X2-255 + WD Green 1.5tb hdd last night, didn't want to bother with the Sempron after all. Idle load is 36 watts with a cheap power supply and 0 tweaking other than defaults. Very happy with that, going to see if i can tweak it down further. Playing just about everything ican throw at it with dxva enabled in ffdshow (shark007 codec pack) it runs about 10-15% cpu load ~42-50 watts by killawatt measure.

Netflix is still a disaster.. nearly 100%cpu load with SL5 installed
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Home Theater Computers
AVS › AVS Forum › Video Components › Home Theater Computers › Sandy Bridge vs. MPC-HC