CPU Performance

The big news with Tegra 3 is that you get four ARM Cortex A9 cores with NEON support instead of just two (sans NEON) in the case of the Tegra 2 or most other smartphone class SoCs. In the short period of time I had to test the tablet I couldn't draw many definitive conclusions but I did come away with some observations.

Linpack showed us healthy gains over Tegra 2 thanks to full NEON support in Tegra 3:

Linpack - Single-threaded

Linpack - Multi-threaded

As expected, finding applications and usage models to task all four cores is pretty difficult. That being said, it's not hard to use the tablet in such a way that you do stress more than two cores. You won't see 100% CPU utilization across all four cores, but there will be a tangible benefit to having more than two. Whether or not the benefit is worth the cost in die area is irrelevant, it only means that NVIDIA (and/or its partners) have to pay more as the price of the end product to you is already pretty much capped.

SunSpider JavaScript Benchmark 0.9.1

Rightware BrowserMark

The bigger benefit I saw to having four cores vs. two is that you're pretty much never CPU limited in anything you do when multitasking. Per core performance can always go up but I found myself bound either by the broken WiFi or NAND speed. In fact, the only thing that would bring the Prime to a halt was if I happened to be doing a lot of writing to NAND over USB. Keyboard and touch interrupts were a low priority at that point, something I hope to see addressed as we are finally entering the era of performance good enough to bring on some I/O crushing multitasking workloads.

Despite having many cores at its disposal, NVIDIA appears to have erred on the side of caution when it comes to power consumption. While I often saw the third and fourth cores fire up when browsing the web or just using the tablet, NVIDIA did a good job of powering them down when their help wasn't needed. Furthermore, NVIDIA also seems to prefer running more cores at lower voltage/frequency settings than fewer cores at a higher point in the v/f curve. This makes sense given the non-linear relationship between voltage and power.

From a die area perspective I'm not entirely sure having four (technically, five) A9 cores is the best way to deliver high performance, but without a new microprocessor architecture it's surely more efficient than just ratcheting up clock speed. I plan on providing a more thorough look at Tegra 3 SoC performance as I spend more time with a fixed Prime, but my initial impressions are that the CPU performance isn't really holding the platform back.

A Lesson in How Not to Launch a Product Tegra 3 GPU: Making Honeycomb Buttery Smooth
POST A COMMENT

204 Comments

View All Comments

  • metafor - Friday, December 02, 2011 - link

    1. The advantages of a companion core apply just as much to single, dual or quad core systems. In each case, individual cores can be power-gated. The companion core is there to provide lower idle power even beyond a single core. So no, going with a dual-core doesn't somehow make a companion core less necessary.

    2. A15 is huge compared to A9. Huge. Both in area and power. If anything, an A15 SoC needs a companion core even more than anything based on A9.

    3. Because A15 is huge, a quad-core in a smartphone form factor isn't very feasible at 32nm. Nor is quad-core really all that useful for the vast majority of use-cases anyway. Especially since A15 performs so much better per-core than an A9.
    Reply
  • phantom505 - Thursday, December 01, 2011 - link

    Yeah, who can put up with a mere 9 hours of continuous playback. It's so bad....

    Oh wait...
    Reply
  • medi01 - Thursday, December 01, 2011 - link

    iPad 25 Wh battery.
    Galaxy Tab 14.8 Wh battery.

    ;)
    Reply
  • thunng8 - Thursday, December 01, 2011 - link

    Why are you comparing the ipad to the 7" Galaxy tab? Of course the ipad will have a bigger battery. Reply
  • quiksilvr - Thursday, December 01, 2011 - link

    I was actually surprised too. To be fair, the A5 chip is huuuuge in the iPad, but that doesn't change the fact that it's still beating a quad core setup.

    My guess is ICS will optimize quad core capabilities more and we'll be seeing a very different picture once that is released.

    So I say wait for ICS and then pass judgement.
    Reply
  • vision33r - Thursday, December 01, 2011 - link

    How is it a surprise? Core i5 can match up against Core i7 in 95% if apps without needing the extra cores.

    In real world test, the difference between Core i5 and i7 performance is hardly measurable.
    Reply
  • name99 - Thursday, December 01, 2011 - link

    Of course by the time ICS is actually available on these sorts of devices, iPad3 with A6 will probably be out...

    Point is: a "my vaporware can beat up your vaporware" contest is generally not very enlightening to anyone.
    Reply
  • daveloft - Thursday, December 01, 2011 - link

    It has nothing to do with the CPU, it's all about the GPU. Reply
  • medi01 - Thursday, December 01, 2011 - link

    Yeah, iPad wins hands down and it's very practical too.
    Think about encoding video in a browser using javascript, for instance
    Reply
  • metafor - Thursday, December 01, 2011 - link

    I think a lot of people (including myself) go with Android because we like the additional features provided by the OS -- true multitasking, choice of a plethora of browsers, third party players, no iTunes, Google integration, etc.

    But yes, iPad 2 has been king of both performance and battery life for a while now.
    Reply

Log in

Don't have an account? Sign up now