NVIDIA Tegra 4 Architecture Deep Dive, Plus Tegra 4i, Icera i500 & Phoenix Hands On
by Anand Lal Shimpi & Brian Klug on February 24, 2013 3:00 PM ESTEver since NVIDIA arrived on the SoC scene, it has done a great job of introducing its ultra mobile SoCs. Tegra 2 and 3 were both introduced with a healthy amount of detail and the sort of collateral we expect to see from any PC silicon vendor. While the rest of the mobile space is slowly playing catchup, NVIDIA continued the trend with its Tegra 4 and Tegra 4i architecture disclosure.
Since Tegra 4i is a bit further out, much of NVIDIA’s focus for today’s disclosure focused on its flagship Tegra 4 SoC due to begin shipping in Q2 of this year along with the NVIDIA i500 baseband. At a high level you’re looking at a quad-core ARM Cortex A15 (plus fifth A15 companion core) and a 72-core GeForce GPU. To understand Tegra 4 at a lower level, we’ll dive into the individual blocks beginning, as usual with the CPU.
ARM’s Cortex A15 and Power Consumption
Tegra 4’s CPU complex sees a significant improvement over Tegra 3. Despite being an ARM architecture licensee, NVIDIA once again licensed a complete processor from ARM rather than designing its own core. I do fundamentally believe that NVIDIA will go the full custom route eventually (see: Project Denver), but that’s a goal that will take time to come to fruition.
In the case of Tegra 4, NVIDIA chose to license ARM’s Cortex A15 - the only vanilla ARM core presently offered that can deliver higher performance than a Cortex A9.
Samsung recently disclosed details about its Cortex A15 implementation compared to the Cortex A7, a similarly performing but more power efficient alternative to the A9. In its ISSCC paper on the topic Samsung noted that the Cortex A15 offered up to 3x the performance of the Cortex A7, at 4x the area and 6x the power consumption. It’s a tremendous performance advantage for sure, but it comes at a great cost to area and power consumption. The area side isn’t as important as NVIDIA has to eat that cost, but power consumption is a valid concern.
To ease fears about power consumption, NVIDIA provided the following data:
The table above is a bit confusing so let me explain. In the first row NVIDIA is showing that it has configured the Tegra 3 and 4 platforms to deliver the same SPECint_base 2000 performance. SPECint is a well respected CPU benchmark that stresses everything from the CPU core to the memory interface. The int at the end of the name implies that we’re looking at purely single threaded integer performance.
The second row shows us the SPECint per watt of the Tegra 3/4 CPU subsystem, when running at the frequencies required to deliver a SPECint score of 520. By itself this doesn’t tell us a whole lot, but we can use this data to get some actual power numbers.
At the same performance level, Tegra 4 operates at 40% lower power than Tegra 3. The comparison is unfortunately not quite apples to apples as we’re artificially limiting Tegra 4’s peak clock speed, while running Tegra 3 at its highest, most power hungry state. The clocks in question are 1.6GHz for Tegra 3 and 825MHz for Tegra 4. Running at lower clocks allows you to run at a lower voltage, which results in much lower power consumption. In other words, NVIDIA’s comparison is useful but skewed in favor of Tegra 4.
What this data does tell us however is exactly how NVIDIA plans on getting Tegra 4 into a phone: by aggressively limiting frequency. If a Cortex A15 at 825MHz delivers identical performance at a lower power compared to a 40nm Cortex A9 at 1.6GHz, it’s likely possible to deliver a marginal performance boost without breaking the power bank.
That 825MHz mark ends up being an important number, because that’s where the fifth companion Cortex A15 tops out at. I suspect that in a phone configuration NVIDIA might keep everything running on the companion core for as long as possible, which would address my fears about typical power consumption in a phone. Peak power consumption is still going to be a problem I think.
Facebook
Twitter
RSS
75 Comments
View All Comments
xsacha - Saturday, March 23, 2013 - link
Tegra4i uses Cortex-A9. Krait is similar to Cortex-A15. The Krait obviously uses way more power and gives way more performance clock-for-clock. So you are comparing apples and oranges here. The 1.9GHz Krait quad-core is roughly equivalent to 2.5GHz+ in a Tegra 4i.name99 - Monday, February 25, 2013 - link
"But in favor of quad-core: software might start using cores a little more effectively w/time--Google and Apple are apparently trying to make WebKit able to do things like HTML parsing and JavaScript garbage collection in the background, and Microsoft's browser team backgrounds JavaScript compilation"It would be wise to design for the technology we have today, not the dream of technology we may one day have. As I have stated elsewhere, there is ample evidence that on the desktop, even today, multiple threads running on more than two cores at once is very rare. (More precisely
- many apps are multithreaded, but those threads tend to be mostly async IO type threads, mostly waiting
- there is a mild win to having three cores available, but it's not much advantage over two cores
- the situation has improved a little over ten years ago (when the first SMT P4s first started appearing) and when there was little advantage to two cores over one. But most of the improvement is the result of OS vendors moving as much stuff as possible of what they do (GUI, IO, etc) onto the second core.)
The only real code that utilizes multiple cores is video-encoding. In particular both games and photo processing do not use nearly as much multi-core as people imagine.
The situation for mobile is the same, only a little worse because there is less of simultaneous heavyweight apps running.
Given these facts, and the way code is actually structured today, 4 cores makes very little sense.
SMT makes sense, mainly in that its power and area footprint is very low, so it's a win on those occasions when the OS can make use of it. Beyond that, if you have excess transistors available, beefed up vectors (wider registers, and wider units) probably makes more sense. You'll notice that these recommendations parallel what Intel has done over the past few years --- they are not idiots, and desktop code is very similar to mobile code.
As for parallel web browsing, people have been publishing about it for years now; but the real world results remain unimpressive. It remains an unfortunate fact that the things that have been converted to parallel don't seem to be, for most sites, the things that are actually gating performance. A similar problem exists with PDF display (still not as snappy as I would like on an iPad3) --- the simple and obvious things you can imagine for parallelizing the rendering aren't the things that are usually the problem.
In both cases, the ideal situation would be to restart with totally redesigned file formats that are non-serial in nature; but that seems to be a "boil-the-ocean" strategy that no-one wants to commit to yet. (Though it would be nice if Apple and Adobe could get together to redefine a PDF2.0 file format that was explicitly parallel, and that seems rather easier than fixing the web.)
Krysto - Sunday, February 24, 2013 - link
It seems Nvidia really pulled off making Tegra 4's GPU 6x faster than Tegra 3, and with 5 Cortex A15 cores and 6x more GPU cores, all in the same size. Pretty impressive. But still quite disappointing for lack of OpenGL ES 3.0 and OpenCL support. I really hope they plan on supporting them in Tegra 5 along with the new 64 CPU and Maxwell-based GPU cores.Mike1111 - Sunday, February 24, 2013 - link
I would really like to see an analysis/comparison of companion core (Nvidia) vs. big.LITTLE (Samsung).lmcd - Sunday, February 24, 2013 - link
BIG.little (fixed it for ARM) isn't even in reference device stage yet is it?Krysto - Monday, February 25, 2013 - link
No need to fix it. The "opposite" style naming is intentional. It's ironic. Get it?phoenix_rizzen - Monday, February 25, 2013 - link
Exynos 5 Octa, which is A15/A7 big.LITTLE, has been demoed. Tegra 4, which is A15 plus a companion core, has been demoed.Neither are commercially available, neither are in shipping products, neither are available to consumers.
IOW, the Cortex-A15 variations for bit.LITTLE have passed the reference stage, and are in the "find companies to use them to build devices" stage. They'll be in consumers' grubby little hands before Christmas 2013.
tviceman - Sunday, February 24, 2013 - link
GPU performance ended up better than I thought it would after the subdued announcement and leaked early prototype benchmarks. Good to see.wongwarren - Monday, February 25, 2013 - link
I wonder which is faster. This or the Snapdragon 600.varad - Monday, February 25, 2013 - link
Snapdragon 600:http://www.anandtech.com/show/6792/lg-optimus-g-pr...
Tegra 4:
http://www.anandtech.com/show/6787/nvidia-tegra-4-...
So if the metric is simply raw performance [since you asked "faster"], looks like the Tegra 4 will win easily against the Snapdragon 600.
A better/fair comparison would be when we have performance numbers for Snapdragon 600 in a tablet or Tegra 4 in a phone.