Sign in with
Sign up | Sign in

Quick Sync: Power Consumption

Core i7-2820QM: Sandy Bridge Shines In Notebooks
By

In order to get a closer look at Quick Sync's benefits in the mobile space, we're using an unprotected 1080p BDAV source on our hard drive to transcode with Cyberlink's MediaEspresso 6.5. If you use MediaEspresso on a Sandy Bridge system, Cyberlink always implements the encoder that Intel provides with MediaSDK, which happens to be the company's turn-key reference library for hardware-accelerated encoding and decoding. As a result, Cyberlink provides two pathways for trascoding: performance or quality. This setting is independent of the overall bitrate target, but it does change the granular aspect of variable bit rate control. Using the performance pathway, scenes with high motion won't be preserved as well, nor will scenes with high detail. But we deal with that in an upcoming article on transcoding image quality.

Today, we are more interested in the Core i7-2820QM, and we want to point out that the reported power values are more than just averages. They are the cumulative power numbers reported on the DC circuit. We have integrated the values over time, but are only reporting the values over a two second interval for charting purposes. These numbers represent the energy required to power the notebook during a transcode workload, with the display set to 100 nits, Wi-Fi off, and in Balanced power mode. So, we are specifically isolating the power consumption of the SSD, LCD panel, motherboard, and CPU.

Using the Performance setting, Quick Sync expectantly uses the least amount of power. Even though it ramps up, it completes the transcode in a fraction of the time. CPU utilization is less impacted by whether you're using hardware or software-based decoding, but going that route results in the transcoding workload taking longer than hardware-based decoding.

When you ratchet up to the Quality setting, there is a larger performance difference between using a hardware and software encoder. Intel's software decoder is fairly efficient, as you never see total CPU usage go beyond 30%. Even though a hardware decoder can make for better transcoding performance and lower power consumption, the biggest gain in both areas comes from using the fixed-function encoder. That is why performance falls largely into two tiers. The encoders are the lynch pin. That is why a decoder matters less in a transcode. If you look at the CPU usage and power profiles, hardware or software encoding, regardless of the decoder used, fall close together.

MediaEspresso 6.5 Transcode
mm:ss
Performance
Quality
Hardware-Based Encode/Decode
Time: 00:48
Avg CPU %: 15.1
Power (W): 35.6
Time: 00:44
Avg CPU %: 11.3
Power (W):39.9
Hardware Encode/Software Decode
Time: 2:51
Avg CPU %: 23.2
Power (W): 31.8
Time: 2:49
Avg CPU %: 22.8
Power (W): 35.0
Software Encode/Hardware Decode
Time: 1:18
Avg CPU %: 51.7
Power (W): 59.4
Time: 2:27
Avg CPU %: 75.8
Power (W): 63.8
Software-Based Encode/Decode
Time: 2:32
Avg CPU %: 52.7
Power (W): 56.1
Time: 3:30
Avg CPU %: 80.6
Power (W): 59.1


When you break the results down into a table, you see some interesting trends. Generally, you can split everything up into a software or hardware encoder camp. Surprisingly, using the quality setting, it takes less time to transcode using the hardware-based decode and encode than using the performance-oriented pathway. Part of this is because the fixed-function encoder is optimized for transcode processing. By preserving the quality, the pipeline is shorter. The software-based decoder doesn't really have a choice, one way or another. It just has to get through the work by brute force CPU horsepower.

Whether you are trying to pack movies onto your iPad for a flight or break down a file to a more manageable upload to YouTube, Quick Sync is the way to go, when it's available. You can do it fast with a low cost to power consumption, and that's a real win when you're on battery power. While it's a shame that Quick Sync isn't going to be available to anyone on the desktop with discrete graphics (at least not yet), you're much more likely to have access to it in the mobile space. Naturally, that's why we wanted to dedicate some time to testing its power impact more thoroughly here.

Ask a Category Expert

Create a new thread in the Reviews comments forum about this subject

Example: Notebook, Android, SSD hard drive

Display all 71 comments.
This thread is closed for comments
  • -3 Hide
    cmartin011 , January 28, 2011 5:40 AM
    Second!!! really a thousand dollars for a mobile cpu
  • 0 Hide
    one-shot , January 28, 2011 5:53 AM
    What are the numbers for battery life for idle, surfing the web, and watching HD video? Several reputable sites have posted up numbers and I'm not seeing a chart that states these numbers, just lots of performance numbers to reiterate the obvious that it's more powerful and more efficient than Arrandale CPUs.
  • -1 Hide
    acku , January 28, 2011 5:59 AM
    This isn't a production notebook so battery life pertaining to this specific notebook is rather pointless in relation to other models. There are other factors at play: LCD panel, battery density, etc... However, platform power consumption numbers are posted on the second to last and last page.

    Andrew
    TomsHardware
  • 2 Hide
    one-shot , January 28, 2011 6:04 AM
    ackuThis isn't a production notebook so battery life pertaining to this specific notebook is rather pointless in relation to other models. There are other factors at play: LCD panel, battery density, etc... However, platform power consumption numbers are posted on the second to last and last page.AndrewTomsHardware


    That isn't what I was looking for. On Anandtech and Tech Report, a Compal notebook with a Core i7 2820QM achieved between six and seven hours of battery life when web browsing. I was looking for a comparison to help me make a more informed decision.

    Something like these is what I was referring to.

    http://www.anandtech.com/show/4084/intels-sandy-bridge-upheaval-in-the-mobile-landscape/9

    http://techreport.com/articles.x/20294/8

    Battery life is not pointless in any way. A pre-production model or not, it's relevant. If helps give us, the consumers, a better perspective to how laptops with these CPUs will perform with regards to battery life.

    I'm surprised it wasn't included.
  • 3 Hide
    acku , January 28, 2011 6:14 AM
    Fair point and I completely agree that battery life is not pointless. But on our point, we did go over power as far as browsing and Flash video if you read our conclusion.

    On an platform level, you can expect a new Sandy Bridge Core i7 to achieve roughly double the battery life of a notebook with an Arrandale Core i5.

    What I disliked about the previous benchmarks (including the ones you referenced) was that they automatically handicapped the benchmark against the Sandy Bridge mobile platform. Forget the whole DTR argument. A 17.3" panel will generally consume more power than a 15.6" (Look at the notebooks it was compared against.) When you isolate it down to the platform level then you can say all-else-being-equal (LCD, hard drive, wireless card, etc...), a notebook based on a Sandy Bridge mobile processor will ~ double battery life. Those other sites showed a roughly 33% improvement because of the other variables at play.

    Remember though that when you are talking about H.264 playback, this is all run through the hardware decoder. You are getting very little battery burn no matter what hardware you are running. What really matters then is the total platforms power consumption and the density of your battery (2.6AH vs 2.9AH cells).

    But back to your main point, if that is what you want to see on a DTR, then we will include it next time. Frankly, I'm more interested in the battery life of non-DTR mobile CPUs. "Normally" people don't care about battery life on a 17.3" mobile workstation.
  • 0 Hide
    bearclaw99 , January 28, 2011 6:35 AM
    Damnnnnn...those are some amazing benches for a laptop CPU. Beats some of the desktop i7s and probably all of AMDs desktop chips
  • 1 Hide
    SteelCity1981 , January 28, 2011 7:05 AM
    Quote:
    If AMD is paying attention, it needs to get its act in order. Brazos is one step up from being a pawn in the AMD Fusion chess set.


    AMD's Brazos platform is very impressive especially the E-350 series that's paired with an Radeon HD 6310 in gaming performance. Soo impressive in fact that the gaming performance rivals that of Core i5 661 in a lot of games and even goes toe to toe with the Core i5 2500k in some games like Call OF Duty Black Ops! As show here....

    http://www.anandtech.com/show/4134/the-brazos-review-amds-e350-supplants-ion-for-miniitx/5
  • -1 Hide
    _Pez_ , January 28, 2011 7:38 AM
    amd is losing ground.. they are taking too long releasing new products.. Intel is expensive.. damm!
  • 0 Hide
    hardcore_gamer , January 28, 2011 7:55 AM
    I think bulldozer will be able to compete in terms of TDP because of the two integer units / core
  • 0 Hide
    Vadim_79 , January 28, 2011 8:56 AM
    I just bought myself an Asus N53SV a couple of days ago, so far it's been great, it can handle any game i throw at it due to the combined intel 3000 and gf540m. Whenever i use the notebook for things like surfing the web it uses the intel 3000, so i get better battery life. I game with the notebook plugged in and set to maximum performance on a 42 inch plasma through hdmi. And it beats my desktop as far as framerates are concerned
  • 1 Hide
    acku , January 28, 2011 9:26 AM
    Quote:
    Quote:
    If AMD is paying attention, it needs to get its act in order. Brazos is one step up from being a pawn in the AMD Fusion chess set.


    AMD's Brazos platform is very impressive especially the E-350 series that's paired with an Radeon HD 6310 in gaming performance. Soo impressive in fact that the gaming performance rivals that of Core i5 661 in a lot of games and even goes toe to toe with the Core i5 2500k in some games like Call OF Duty Black Ops! As show here....

    http://www.anandtech.com/show/4134/the-brazos-review-amds-e350-supplants-ion-for-miniitx/5



    Do you really want to play Call of Duty Black Ops at 1024 x 768 at low quality? I wouldn't ever want to punish any TomsHardware reader that harshly. :) 

    Andrew
    TomsHardware
  • 1 Hide
    silverblue , January 28, 2011 9:59 AM
    I'm really not sure that's the intention. Sure, it supports DirectX 11, but we all know that below the 5700 series, there's little point using it. The true strength of Brazos' GPU is slightly older games or ones that don't require masses of bandwidth because that single channel memory interface will strangle it in the end.
  • 1 Hide
    acku , January 28, 2011 10:02 AM
    Quote:
    I'm really not sure that's the intention. Sure, it supports DirectX 11, but we all know that below the 5700 series, there's little point using it. The true strength of Brazos' GPU is slightly older games or ones that don't require masses of bandwidth because that single channel memory interface will strangle it in the end.


    I think you hit the nail on its head right there. AMD never really positioned the Brazos platform as a "gaming platform." It can't handle it. It works better as a more powerful enhancement over an Atom.

    Same thing goes for the HD Graphics 3000. If you have a DTR, it is likely you are going to get a discrete chipset anyways.
  • 0 Hide
    silverblue , January 28, 2011 10:11 AM
    It can play modern games without textures flickering all over the place, but most won't be too smooth, and you certainly wouldn't want to enable AA thanks to the bandwidth issue unless you're really limited by CPU performance (and even then...).

    It's fantastic that you can play some modern games in low detail with a decent framerate with something as small as your fingernail and use very little power doing it, but people shouldn't get their hopes up that this is, say, a console killer. Let's wait and see what Enhanced Bobcat is like for that sort of thing.
  • 0 Hide
    marraco , January 28, 2011 10:16 AM
    The 7-zip chart needs to be fixed.
  • 0 Hide
    Onus , January 28, 2011 10:54 AM
    On a lighter note, "RLUMark" is unpronounceable. I would like to suggest "IRLMark," (pronounced "Earl-Mark") for "In-Real-Life" for your realistic benchmark.
  • 1 Hide
    SteelCity1981 , January 28, 2011 11:13 AM
    Quote:
    Do you really want to play Call of Duty Black Ops at 1024 x 768 at low quality? I wouldn't ever want to punish any TomsHardware reader that harshly.


    Well the same can be said when benchmarking an Intel Core i7/5/3 2xxx using its integraded graphics unit on game like Call Of Duty Black Opts. But I think you are missing the point in regards to the AMD's E-350 and the Call of Duty Black Ops benchmark. Of course no one is really going to play a game like Call Of Duty Black Opts on a integrated graphics unit with everything on low settings unless they are really that desperate. The point of doing that benchmark was to show the capabilities of the E-350's integrated graphics unit and for a processor aimed at the budget and ultra budget markets that can compete with mid-range processors with integrated graphics units built in, that's pretty impressive to say the least.
  • -1 Hide
    acku , January 28, 2011 11:50 AM
    SteelCity1981Well the same can be said when benchmarking an Intel Core i7/5/3 2xxx using its integraded graphics unit on game like Call Of Duty Black Opts. But I think you are missing the point in regards to the AMD's E-350 and the Call of Duty Black Ops benchmark. Of course no one is really going to play a game like Call Of Duty Black Opts on a integrated graphics unit with everything on low settings unless they are really that desperate. The point of doing that benchmark was to show the capabilities of the E-350's integrated graphics unit and for a processor aimed at the budget and ultra budget markets that can compete with mid-range processors with integrated graphics units built in, that's pretty impressive to say the least.


    I'll agree with that sentiment. However, simply stating that it the E-350 can perform similarly still doesn't address how similar it is in higher resolutions or a realistic quality setting. Or even in real life. It's a different market altogether. As Chris has often said, "Sorry your princess is in a different castle."

    That said, he actually covered all of this in his original desktop Brazos review. And it isn't right in my mind to make that type of comparison anyways. On the i7-2820QM, it's a mobile CPU that is going into a DTR notebook and is almost guaranteed to have discrete chip. On the mobile side things come out as systems, rarely do you get to simply pick and choose CPU + Graphics. What is the point about talking about the graphic short comings on this CPU when it is certainly always going to be paired with a powerful GPU?

    With the E-350, you are talking about nettops and netbooks. You won't be able to game anything larger than 1366 x 768 even at the most optimistic notebook configuration. So 1024 x 768 is a reasonable expectation given that is the resolution most often seen on the netbook side. Remember, AMD is hitting low prices with their 100 CPU/mobo combo, so this it is truly meant as a budget option.

    Meanwhile the i7-2820QM is certainly always going to be in a 15.6" LCD system or larger. Brazos is $500 and under. That is the target. With the i5-2820QM you are looking at systems that are going to be priced at least $1,000 plus its going to come with a discrete chip. Realistically, we are talking about at least $1,500. The i7-2630 is down the ladder on the Asus N53SV and probably ran around 1k, but I'm sure vadim_79 can jump and share the final price tag.
  • 0 Hide
    acku , January 28, 2011 11:54 AM
    Quote:
    It can play modern games without textures flickering all over the place, but most won't be too smooth, and you certainly wouldn't want to enable AA thanks to the bandwidth issue unless you're really limited by CPU performance (and even then...).

    It's fantastic that you can play some modern games in low detail with a decent framerate with something as small as your fingernail and use very little power doing it, but people shouldn't get their hopes up that this is, say, a console killer. Let's wait and see what Enhanced Bobcat is like for that sort of thing.


    That Enhanced Bobcat will be a 2012 move. And that would be "some modern games." CodBO isn't DX11. And I doubt anyone wants to attempt to play Crysis on a Brazos system even at 1024x768

    http://www.tomshardware.com/reviews/amd-fusion-brazos-performance,2790-7.html

    You will be wanting to look toward Krishna. Ontario and Wichita will still be ala Atom flavors.

    AMD's mobile plans hang on Llano and the Sabine platform. I'm teething to see them in action.
  • 0 Hide
    acku , January 28, 2011 11:55 AM
    Quote:
    On a lighter note, "RLUMark" is unpronounceable. I would like to suggest "IRLMark," (pronounced "Earl-Mark") for "In-Real-Life" for your realistic benchmark.

    I may just have to steal that idea! Thanks
Display more comments