AMD's Trinity : An HTPC Perspective
by Ganesh T S on September 27, 2012 11:00 AM ESTAMD provided us with an A10-5800K APU along with the Asus F2 A85-M Pro motherboard for our test drive. Purists might balk at the idea of an overclockable 100W TDP processor being used in tests intended to analyze the HTPC capabilities. However, the A10-5800K comes with the AMD Radeon HD 7660D, the highest end GPU in the Trinity lineup. Using this as the review platform gives readers an understanding of the maximum HTPC capabilities of the Trinity lineup.
The table below presents the hardware components of our Trinity HTPC testbed:
| Trinity HTPC Testbed Setup | |
| Processor | AMD A10-5800K - 3.80 GHz (Turbo to 4.2 GHz) |
| AMD Radeon HD 7660D - 800 MHz | |
| Motherboard | Asus F2A85-M Pro uATX |
| OS Drive | OCZ Vertex2 120 GB |
| Memory | G.SKILL Ares Series 8GB (2 x 4GB) SDRAM DDR3 2133 (PC3 17000) F3-2133C9Q-16GAB CAS 9-11 -10-28 2N |
| Optical Drives | ASUS 8X Blu-ray Drive Model BC-08B1ST |
| Case | Antec Skeleton ATX Open Air Case |
| Power Supply | Antec VP-450 450W ATX |
| Operating System | Windows 7 Ultimate x64 SP1 |
| Display / AVR | Acer H243H / Pioneer Elite VSX-32 + Sony Bravia KDL46EX720 |
| . | |
The Trinity platform officially supports DDR3-1866 modules. Towards this, we obtained a 16 GB DDR3-2133 Ares kit from G.Skill for our testbed. Using this kit made it possible to study HTPC behaviour from a memory bandwidth perspective.
The software setup for the Trinity HTPC testbed involved the following:
| Trinity HTPC Testbed Software Setup | |
| Blu-ray Playback Software | CyberLink PowerDVD 12 |
| Media Player | MPC-HC v1.6.3.5818 |
| Splitter / Decoder | LAV Filters 0.51.3 |
| Renderers | EVR / EVR-CP (integrated in MPC-HC v1.6.3.5818) |
| madVR v0.83.4 | |
The madVR renderer settings were fixed as below for testing purposes:
- Decoding features disabled
-
Deinterlacing set to:
- automatically activated when needed (activate when in doubt)
- automatic source type detection (i.e, disable automatic source type detection is left unchecked)
- only look at pixels in the frame center
- be performed in a separate thread
-
Scaling algorithms were set as below:
- Chroma upscaling set to SoftCubic with softness of 100
- Luma upscaling set to Lanczos with 4 taps
- Luma downscaling set to Lanczos with 4 taps
-
Rendering parameters were set as below:
- Start of playback (including post-seek) was delayed till the render queue filled up
- Automatic fullscreen exclusive mode was used
- A separate device was used presentation, and D3D11 was used
- CPU and GPU queue sizes were set to 32 and 24 respectively
- Under exclusive mode settings, the seek bar was enabled, switch to exclusive mode from windowed mode was delayed by 3 seconds and 16 frames were configured to be presented in advance. The GPU was set to fush after the intermediate render steps, copy to back buffer and after D3D peresentation. In addition, the GPU was set to wait (sleep) after the last render step.
Unlike our Ivy Bridge setup, we found the windowed mode to be generally bad in terms of performance compared to exclusive mode. Also, none of the options to trade quality for performance were checked.
Facebook
Twitter
RSS
49 Comments
View All Comments
ganeshts - Thursday, September 27, 2012 - link
Hmmm.. all vendors tag 23.976 Hz as 23 Hz in the monitor / GPU control panel settings. So, when I set the panel to 23 Hz, I am actually expecting 23.976 Hz. However, this platform gives me 23.977 Hz which is a departure from the usually accurate AMD cards that I have seen so far.ChronoReverse - Thursday, September 27, 2012 - link
23.977 and 23.976 are so close that it's basically the same (the error in measuring tools would be as large as the difference). I'd only be concerned if it were 23.970.In any case, from looking at the screenshots in the gallery, the only frequency looking rather off is 60Hz (although my AMD card has always given similar lower than 60Hz results anyway).
ganeshts - Thursday, September 27, 2012 - link
Note that these are in Hz, not MHz. So, the margin for error is quite large. In fact, madVR statistics deliver accurate refresh rates up to 6 decimal digits (as the screenshots show).To read more on why the 0.001 Hz difference matters for SOME people, look this up: http://www.anandtech.com/show/4380/discrete-htpc-g...
In short, with the 0.001 Hz difference, the renderer might need to repeat a frame every ~17 minutes. I am NOT saying that this is a serious issue for everyone, but there are some readers who do care about this (as evidenced by the range of opinions expressed in this thread: http://www.avsforum.com/t/1333324/lets-set-this-st...
ChronoReverse - Thursday, September 27, 2012 - link
That thread on avsforum is talking about 24FPS playback where if you got 23.97x instead, it's a stutter about every 42 seconds which is terrible and clearly not acceptable (to my eye anyway).Still, I do admit that even a single stutter every 17 minutes is noticeable.
Also, I had misread that part of the review a bit since for some reason I had the impression it was saying the performance of AMD has diminished when it's still about the same +/- 0.002Hz
jeremyshaw - Thursday, September 27, 2012 - link
Wasn't AMD's first APU Brazos, not Llano? Or was it too small to really count!?ganeshts - Thursday, September 27, 2012 - link
Technically correct, but it didn't compete in the same level as the Clarkdales / Arrandales / Sandy Bridge lineup :)jamawass - Thursday, September 27, 2012 - link
"The video industry is pushing 4K and it makes more sense to a lot of people compared to the 3D push. 4K will see a much faster rate of adoption compared to 3D, but Trinity seems to have missed the boat here. AMD's Southern Islands as well as NVIDIA's Kepler GPUs support 4K output over HDMI, but Trinity doesn't have 4K video decode acceleration or 4K display output over HDMI."Although this statement is technically correct it has no real world relevance. At this time people who can afford 4k TVs ( if there any commercially available ones at this time) won't be messing around with cheap htpcs. It's an inconsequential statement made just to detract from AMD's overall superiority with this product in the htpc market.
If I was in AMD's shoes why would I dedicate resources to a nonexistent market ? Has anyone actually tested Nvidia or Intel's 4K output over HDMI to see whether they actually work? In the early days of HDCP all the video card manufacturers were claiming compliance but real world compatibility was a different matter.
'
ganeshts - Thursday, September 27, 2012 - link
I had the same caveat in the Ivy Bridge HTPC review. Surprised you didn't notice that, but you notice this :) Ivy Bridge doesn't support 4K over HDMI yet.Anyways, yes, we have test 4K output from both NV and AMD. When AMD 7750 was released, we didn't have access to a 4K display, but things changed when the GT 640 was released:
http://www.anandtech.com/show/5969/zotac-geforce-g...
I don't have a sample image ready for the 7750, but I can assure you that it works as well as NVIDIA's and I have personally tested it. In fact, AMD was the first to 4K output over HDMI.
JNo - Saturday, September 29, 2012 - link
More importantly do you have any 4K films to watch? No. Will you in the immediate future? No. Even then, when will *most* new films coming out be available in 4K? Probably in 5 years time when you'd build a new HTPC anyway.The 4K thing is absolutely irrelevant at this point (unlike 3D I'd argue because you can go into plenty of shops and buy actual 3D media).
After Hi Def came out hardware (TVs) were available quickly but it took a *long* time before there was plenty of 1080p material anyway (note use of the word, 'plenty'). Hell, most people I know are still watching stuff in SD. Laughably, 4K isn't even close to being out yet, let alone the content.
The whole thing's a red herring right now and for a long while.
Cotita - Thursday, September 27, 2012 - link
I'm not sure I'd go for an A10.Even a A4 3420 would do pretty much the same.
Heck, If I don't care about HD flash or silverlight even a E-350 is enough