Video Encoding Performance

Our DivX test is the same DivX / XMpeg 5.03 test we've run for the past few years now, the 1080p source file is encoded using the unconstrained DivX profile, quality/performance is set balanced at 5 and enhanced multithreading is enabled.

Xmpeg + DivX Encode

Despite the greatness that is Quick Sync, there are no editing/high quality transcode tools that support Intel's hardware transcode engine. Luckily, Sandy Bridge is still very fast when it comes to software encoding. Our WME test only shows minimal gains thanks to the architectural improvements however.

Windows Media Encoder 9 - Advanced Profile

Graysky's x264 HD test uses x264 to encode a 4Mbps 720p MPEG-2 source. The focus here is on quality rather than speed, thus the benchmark uses a 2-pass encode and reports the average frame rate in each pass.

Other than the Core i7 980X, there's nothing quicker than Sandy Bridge. The Core i7 2600K is 10% faster than the Core i7 975, and the 2500K easily outpaces its Lynnfield rivals. The i3 2100 is quicker than its predecessor, however not by much. In these heavily threaded situations, AMD's Athlon II X4 645 is a better option than the 2100.

x264 HD Benchmark - 1st Pass

x264 HD Benchmark - 2nd Pass

x264 HD 3.03 Benchmark - 1st Pass

x264 HD 3.03 Benchmark - 2nd Pass

SYSMark 2007 & Photoshop Performance 3D Rendering Performance
Comments Locked

283 Comments

View All Comments

  • aviat72 - Tuesday, January 4, 2011 - link

    Though SB will be great for some applications, there are still rough edges in terms of the overall platform. I think it will be best to wait for SNB-E or at least the Z68. SNB-E seems to be the best future-proofing bet.

    I also wonder how a part rated for 95W TDP was drawing 111W in the 4.4GHz OC (the Power Consumption Page). SB's power budget controller must be really smart to allow the higher performance without throttling down, assuming your cooling system can manage the thermals.
  • marraco - Tuesday, January 4, 2011 - link

    I wish to know more about this Sandy Bridge "feature":

    http://www.theinquirer.net/inquirer/news/1934536/i...
  • PeterO - Tuesday, January 4, 2011 - link

    Anand, Thanks for the great schooling and deep test results -- something surely representing an enormous amount of time to write, produce, and massage within Intel's bumped-forward official announcement date.

    Here's a crazy work-around question:

    Can I have my Quick Synch cake and eat my Single-monitor-with-Discrete-Graphics-card too if I, say:

    1). set my discrete card output to mirror Sandy Bridge's IGP display output;

    2). and, (should something exist), add some kind of signal loopback adapter to the IGP port to spoof the presence of a monitor? A null modem, of sorts?

    -- I have absolutely no mobo/video signaling background, so my idea may be laugh in my face funny to anybody who does but I figure it's worth a post, if only for your entertainment. :)
  • Hrel - Wednesday, January 5, 2011 - link

    It makes me SO angry when Intel does stupid shit like disable HT on most of their CPU's even though the damn CPU already has it on it, they already paid for. It literally wouldn't cost them ANYTHING to turn HT on those CPU's yet the greedy bastards don't do it.
  • Moizy - Wednesday, January 5, 2011 - link

    The HD Graphics 3000 performance is pretty impressive, but won't be utilized by most. Most who utilize Intel desktop graphics will be using the HD Graphics 2000, which is okay, but I ran back to the AMD Brazos performance review to get some comparisons.

    In Modern Warfare 2, at 1024 x 768, the new Intel HD Graphics 2000 in the Core i3 2100 barely bests the E-350. Hmm--that's when it's coupled with a full-powered, hyper-threaded desktop compute core that would run circles around the compute side of the Brazos E-350, an 18w, ultra-thin chip.

    This either makes Intel's graphics less impressive, or AMD's more impressive. For me, I'm more impressed with the graphics power in the 18w Brazos chip, and I'm very excited by what mainstream Llano desktop chips (65w - 95w) will bring, graphics-wise. Should be the perfect HTPC solution, all on the CPU (ahem, APU, I mean).

    I'm very impressed with Intel's video transcoding, however. Makes CUDA seem...less impressive, like a bunch of whoop-la. Scary what Intel can do when it decides that it cares about doing it.
  • andywuwei - Wednesday, January 5, 2011 - link

    not sure if anybody else noticed. CPU temp of the i5@3.2GHz is ~140 degrees. any idea why it is so high?
  • SantaAna12 - Wednesday, January 5, 2011 - link

    Did I miss the part where you tell of about the DRM built into this chip?
  • Cb422 - Wednesday, January 5, 2011 - link

    When will Sandy Bridge be available on Newegg or Amazon for me to purchase?
  • DesktopMan - Thursday, January 6, 2011 - link

    Very disappointed in the lack of vt-d and txt on k-variants. They are after all the high end products. I also find the fact that only the k-variants having the faster GPU very peculiar, as those are the CPUs most likely to be paired with a discrete GPU.
  • RagingDragon - Thursday, January 6, 2011 - link

    Agreed. I find the exclusion of VT-d particularly irritating: many of the overclockers and enthusiasts to whom the K chips are marketed also use virtualization. Though I don't expect many enthusiasts, if any, to miss TXT (it's more for locked down corporate systems, media appliances, game consoles, etc.).

    With the Z68 chipset coming in the indeterminate near future, the faster GPU on K chips would have made sense if the K chips came with every other feature enabled (i.e. if they were the "do eveything chips").

    Also, I'd like to have the Sandy Bridge video encode/decode features separate from the GPU functionality - i.e. I'd like to choose between Intel and Nvidia/AMD video decode/encode when using a discrete GPU.

Log in

Don't have an account? Sign up now