Power Consumption

Intel has a full process node advantage when you compare Ivy Bridge and Trinity, as a result of that plus an architectural efficiency advantage you just get much better power consumption from the Core i3 than you do with Trinity. Idle power is very good but under heavy CPU load Trinity consumes considerably more power. You're basically looking at quad-core Ivy Bridge levels of power usage under load but performance closer to that of a dual-core Ivy Bridge. AMD really needs a lot of design level efficiency improvements to get power consumption under control. Compared to Llano, Trinity is a bit more efficient it seems so there's an actual improvement there.

Power Consumption - Idle

Power Consumption - Load (x264 HD 5.0.1 2nd Pass)

Of course on the processor graphics side the story is much closer with Trinity being a bit more power hungry than Ivy Bridge, but not nearly by this margin.

Overclocking & Power Consumption Final Words
Comments Locked


View All Comments

  • Aone - Wednesday, October 3, 2012 - link

    Who being in his right buys high-end discrete GPU for cheap CPU or APU?

    Plus, those who buy cheap CPUs usually don't have money for high-end discrete GPU.
  • Gaugamela - Wednesday, October 3, 2012 - link

    Here are benchmarks that test the importance of faster RAM in these APUs. The difference in performance is astonishing. http://hexus.net/tech/reviews/cpu/46073-amd-a10-58...

    By overclocking and using 2133Mhz RAM the A10-5800 can get approximately a 30% increase in 3DMark and some games.
    These Trinity APUs seem to be really interesting to tinker with.
  • creed3020 - Wednesday, October 3, 2012 - link

    Thanks so much for posting that. I've been looking for this exact testing of Trinity. AT did this previously with Llano but forgot this crucial test with Trinity.

    It really helps system builders to set expectations for performance if a client doesn't want to pay for faster memory, or if they do want more performance we can quantify how much an improvement faster memory will have.
  • mikato - Wednesday, October 3, 2012 - link

    Holy moly
  • vozmem - Wednesday, October 3, 2012 - link

    Keep encouraging AMD, guys.
  • rarson - Wednesday, October 3, 2012 - link

    Why in the world did you not mention which video card you were using on this page? I see that it's mentioned in the test bed, but why the heck do I have to go back and check that when you could have easily mentioned it on the discrete test page?

    Also, why are you using a 5870 with this? Who the hell is going to pair a new A8 or A10 Trinity with a 5870? That's completely illogical. Couldn't you have tried something newer, perhaps something within the same architecture? Extremely puzzling.
  • etamin - Wednesday, October 3, 2012 - link

    And why was an FX-8150 thrown into the DISCRETE PROCESSOR GRAPHICS benchmark?
  • Hardcore69 - Wednesday, October 3, 2012 - link

    HA! Glad I went with an i3 3220 for this office box. Look at the power consumption at load, look at the single threaded benchmarks, even look at the multi threaded benchmarks. AMD is crap. It still hasn't caught up. And there are very few upgrade options compared to Intel. If you want to play games, a dedicated GPU is still vastly better. For other basic tasks, FAIL.
  • rarson - Wednesday, October 3, 2012 - link

    You paid more money.

    They're called trade-offs. That's reality.
  • Nil Einne - Friday, October 5, 2012 - link

    Has anyone come across real world power consumption figures for either the A8-5500 vs A8-5600K or the A10-5700 vs A10-5800K. These have different TDPs, 65W vs 100W and slightly different clocks. But I'm wondering whether the K ones are really that bad in general or it's partially that they wanted more headroom since the K ones are to some extent designed to be overclocked. Of course the different ratings means that you may get unlucky and get a fairly high consumption K processor because of binning but still may be relevent. I'm somewhat out of date and not familiar with how turbo works, but I'm guessing the higher binning means it will stay at turbo for longer so a proper test should also try limiting the K to be the same as the non K just to see if that's the primary reason for any differences. (Ideally also limit the frequencies.)

    Most reviews including this one seem to be of the Ks I presume because that's what AMD sent out for testing.


Log in

Don't have an account? Sign up now