Linked by Thom Holwerda on Sun 7th Aug 2011 22:46 UTC
Thread beginning with comment 484142
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[5]: "Best" is in the eye of the beholder
by chithanh on Tue 9th Aug 2011 12:08
in reply to "RE[4]: "Best" is in the eye of the beholder"
"The Llano is underpowered for 1080i using advanced deinterlacers and other video post-processing.
I looked up the Anandtech article on AMD A8-3850 for HTPC (http://www.anandtech.com/print/4479) and it does mention problems with 1080p. So I must have been thinking of something else... though now I'm curious whether that can be fixed in software. "
The suitability of Llano for HTPC is extensively discussed in this AVS Forum thread: http://www.avsforum.com/avs-vb/showthread.php?t=1333378
The Anandtech article's 1080p60 issues are because of one single post-processing feature (dynamic contrast) in combination with Anandtech's use of DDR3-1333 and discussed starting from page 11.
tl;dr -> posts 343 and 353 summarize the situation.
Edited 2011-08-09 12:10 UTC
RE[6]: "Best" is in the eye of the beholder
by ilovebeer on Tue 9th Aug 2011 15:56
in reply to "RE[5]: "Best" is in the eye of the beholder"
The suitability of Llano for HTPC is extensively discussed in this AVS Forum thread: http://www.avsforum.com/avs-vb/showthread.php?t=1333378
The Anandtech article's 1080p60 issues are because of one single post-processing feature (dynamic contrast) in combination with Anandtech's use of DDR3-1333 and discussed starting from page 11.
tl;dr -> posts 343 and 353 summarize the situation.
The Anandtech article's 1080p60 issues are because of one single post-processing feature (dynamic contrast) in combination with Anandtech's use of DDR3-1333 and discussed starting from page 11.
tl;dr -> posts 343 and 353 summarize the situation.
First, the people who are rambling about the Llano being a faster cpu then an Atom is completely absurd. Aside of the obvious problems with those types of statements, when you're talking about an htpc using a hardware decoder, the cpu is sitting IDLE most of the time. The fact you're using a hardware decoder practically removes the cpu from the equation.
So, now you look at the GPU... The speed of the memory, the shaders, the deinterlacers, bla bla bla.. Is the Llano capable of being used in an htpc environment? Yes, of course. Will it provide competition for Nvidia's already proven high end deinterlacers? Absolutely not.
There's no question there are areas where Llano falls short in the htpc arena. That doesn't mean it's a bad choice -- it isn't. It's just not the best. People, wrongly, cling to 1080p performance as the measuring stick.. The real test is how it handles 1080i with post-processing. Anyone that doesn't understand that has a lot to learn about this subject.




Member since:
2008-10-22
I looked up the Anandtech article on AMD A8-3850 for HTPC (http://www.anandtech.com/print/4479) and it does mention problems with 1080p. So I must have been thinking of something else... though now I'm curious whether that can be fixed in software.
The FM1 socket boards from Gigabyte do: http://www.gigabyte.com/products/list.aspx?s=42&jid=10&p=2&v=27
That's usually the best way to buy complete bare bones systems as well.
There's been some confusion on the thread about this. If Thom's looking to replace his motherboard, why not just by a media extender? That's certainly possible but it sounds like he wants to keep as much of his existing hardware as possible. If he's got DDR3 RAM at compatible speeds then that will help toward that goal... (which seems unlikely given the age of the system mentioned in his post)
re SATA port multipliers, I've heard a lot about this on the OpenIndiana mailing list but haven't encountered it yet myself. I had just written it off as someone trying to build a storage system with cheap desktop hardware that wasn't designed with that goal in mind.
-Gary
Edited 2011-08-09 00:41 UTC