Next week you’re going to hear everyone in the mobile industry trumpet NVIDIA’s quad core Tegra 3 as the best thing since sliced bread. Why? Because four cores simply have to better than two cores, right? Texas Instruments, in what can best be described as a public service announcement, just uploaded a video to YouTube showing how their dual core ARM Cortex A15 chip compares to NVIDIA’s quad core ARM Cortex A9 chip on EEMBC’s BrowsingBench testing tool. The benchmark is really easy to understand: 20 web pages get loaded up, one right after the other, and the time it takes to load all of said pages gets recorded. TI’s chip did the test in 95 seconds. NVIDIA’s chip took 201 seconds. In other words, the Texas Instruments OMAP5 is more than twice as fast as a Tegra 3, despite having half the cores. How is this possible? Because processors can’t be measured by how many cores they have or how many gigahertz they can achieve, instead they need to be measured by how many instructions they can execute per clock cycle. In this case, the ARM Cortex A15 is hellishly faster than the ARM Cortex A9.
There’s a catch to all of this of course. The NVIDIA Tegra 3 is going to be shown off next week and will end up in consumer hardware that’ll be on store shelves by the end of the first half of this year. The Texas Instruments OMAP5 however will barely make it into products out by the end of this year, and realistically speaking there’s a good chance it’ll end up in early 2013 devices. By that time we’ll likely hear about NVIDIA’s Tegra 4, currently codenamed “Wayne”, which is rumored to utilize Cortex A15 cores.
Now all this being said, the whole NVIDIA vs. Texas Instruments thing, at the end of the day we’re cheering for Qualcomm. Their dual core Snapdragon S4 was recently benchmarked and it murders everything on the market, including the Tegra 3. Best of all it’ll end up in devices coming out in Q2 of this year!
[Via: AnandTech]