GPU Workload

NVIDIA's only performance advantage on the SoC side compared to Clover Trail at this point is in its GPU. Tegra 3's GPU is faster than the high clocked PowerVR SGX 545 in Clover Trail. While we don't yet have final GPU benchmarks under Windows RT/8 that we can share numbers from, the charts below show power consumption in the same DX title running through roughly the same play path.

NVIDIA's GPU power consumption is more than double the PowerVR SGX 545's here, while its performance advantage isn't anywhere near double. I have heard that Imagination has been building the most power efficient GPUs on the market for quite a while now, this might be the first argument in favor of that heresay.

WebXPRT & TouchXPRT Wireless Web Browsing Battery Life Test
Comments Locked

163 Comments

View All Comments

  • yyrkoon - Tuesday, December 25, 2012 - link

    same OS, different HAL.
  • Krysto - Monday, December 24, 2012 - link

    Good job comparing a last-gen chip like Tegra 3 with Intel's latest Atom chip, Anand. Just because Microsoft is too slow to adopt a cutting edge ARM chip for Windows, doesn't mean Intel is now "toe to toe" with ARM.

    Compare Nexus 10 and iPad 4 with Clover Trail in GPU performance and price, and then let me know how it went.
  • croc - Monday, December 24, 2012 - link

    People that tend to use terms like 'fact' also seem to be such complete tools...
  • Barnassey - Monday, December 24, 2012 - link

    My main thing is why is anand comparing a QUAD core cpu to a DUAL core With hyper-threading cpu? Of course the hyperthreading cpu will use less power.
  • Reikon - Monday, December 24, 2012 - link

    Because the dual core CPU still outperforms the quad core one?
  • Barnassey - Monday, December 24, 2012 - link

    No because the power consumption will be different. Thats the point im trying to make. Plus anand has shown hes a little biased over the years.
  • r1cky4n - Monday, December 24, 2012 - link

    a single atom core is not equivalent to a single a9 core

    The important metric is performance/watt, not watt/core
  • sseemaku - Tuesday, December 25, 2012 - link

    So you want them to disable one of the channels in memory controller and reduce the cache size to match tegra3 for a fair comparison?
  • tempestglen - Monday, December 24, 2012 - link

    It seems that "GPU Workload" and ""Photo Enhance" CPU consumption on summary table are inconsistent with diagrams.
  • tempestglen - Monday, December 24, 2012 - link

    It also seems that Kraken test diagram of total platform power is just same as cpu only diagram. I guess anandtech posted a wrong picture.

    Merry Christmas!

Log in

Don't have an account? Sign up now