HiSilicon Kirin 960: A Closer Look at Performance and Power
by Matt Humrick on March 14, 2017 7:00 AM EST- Posted in
- Smartphones
- Mobile
- SoCs
- HiSilicon
- Cortex A73
- Kirin 960
Final Words
HiSilicon’s Kirin 950 delivered impressive performance and efficiency, raising our expectations for its successor. And on paper at least, the Kirin 960 seems better in every way. It incorporates ARM’s latest IP, including A73 CPUs, the new Mali-G71 GPU with more cores, and a CCI-550 interconnect. It offers other improvements too, such as a new modem that supports higher LTE speeds and UFS 2.1 support. But when it comes to performance and efficiency, the Kirin 960 improves in some areas and regresses in others.
The Kirin 960’s A73 CPU cores are marginally faster than the 950’s A72 cores when handling integer workloads, with a more noticeable lead over Qualcomm’s Kryo and the older A57. When looking at floating-point IPC, the opposite is true, with Qualcomm’s Kryo and Kirin 950’s A72 cores posting better results than the 960’s A73.
Some of this performance regression may be explained by Kirin 960’s memory performance. Both latency and read bandwidth improve for its larger 64KB L1 cache, but write bandwidth is lower than Kirin 950. The 960’s L2 cache bandwidth is also lower for both read and write. Its latency to main memory improves by 25%, however, and bandwidth improves by an impressive 69%.
What’s really disappointing (and puzzling) about Kirin 960, though, is that its CPU efficiency is actually worse than the 950’s. ARM did a lot of work to reduce the A73’s power consumption relative to the A72, but the Kirin 960’s A73 cores see a substantial power increase over the 950’s A72 cores. The poor efficiency numbers are likely a combination of HiSilicon’s specific implementation and the switch to the 16FFC process. This was definitely an unexpected result considering the Mate 9’s excellent battery life. Fortunately, Huawei was able to save power elsewhere, such as the display, to make up for the SoC’s power increase, but it’s difficult not to think about how much better the battery life could have been.
Power consumption for Kirin 960’s GPU is even worse, with peak power numbers that are entirely inappropriate for a smartphone. Part of the problem is poor efficiency, again likely a combination of implementation and process, which is only made worse by an overly aggressive 1037MHz peak operating point that only serves to improve the spec sheet and benchmark results.
The Kirin 960 is difficult to categorize. It’s definitely not a clear upgrade over the 950, but it does just enough things right that we cannot dismiss it outright either. For example, its generally improved integer performance and lower system memory latency give it an advantage over the 950 in many real-world workloads. We cannot completely condemn its GPU either, because its sustained performance, at least in the Mate 9’s large aluminum chassis, is on par with or better than competing flagship phones, as is its battery life when gaming. Certainly the Mate 9 proves that Kirin 960 is a viable flagship SoC as long as Huawei puts in the effort to work around its flaws. But with a new generation of 10nm SoCs just around the corner, those flaws will only become more apparent.
Facebook
Twitter
RSS
86 Comments
View All Comments
nikhilmaurya10 - Friday, July 21, 2017 - link
Believe me I bought the honor 8 pro with kirin 960 at 466 USD(india 30k RS), so they have achieved their goal of making a flagship level chip for below flagship price. After reading this review and finding that 8W power of that GPU i am worried about the VR content on this 2k display. One thing is good that this phone is huge metallic slab, that should keep it somewhat cool.socalbigmike - Thursday, March 16, 2017 - link
They ARE sponsored by Huawei.Meteor2 - Wednesday, March 15, 2017 - link
This. This is a great article but what's missing is the A10 (and Core M and Atom for comparison).I'm less interested in the deeper technical stuff, tbh. But I'm very interested in the performance, power consumption, and resulting efficiency. So I'd love to see this test battery for the A10 and Core too.
Mind you, why don't you do SPi2000 and GB4 against power consumption, rather than only PCMark?
jjj - Tuesday, March 14, 2017 - link
ARM was comparing A73 on 10nm vs A72 on 16nm in efficiency ,not peak power for both on same process.Likely the memory controller and the interconnect have an impact too in increasing the differences between the 950 and 960.
Matt Humrick - Tuesday, March 14, 2017 - link
ARM's power comparison was for the same process and same frequency.jjj - Tuesday, March 14, 2017 - link
On a per same task basis not peak load. http://images.anandtech.com/doci/10347/11.PNGIn this slide it's 10nm vs 16nm http://images.anandtech.com/doci/10347/1_575px.PNG
degasus - Tuesday, March 14, 2017 - link
> I cannot think of any CPU-centric workloads for a phone that would load two big cores for anywhere near this longYou haven't run an emulator, have you? With a bit improved GPU drivers (here, EXT_buffer_storage), this will be a very good device for playing Gamecube and Wii games. This will stress two threads, and a bit the GPU.
tuxRoller - Friday, March 17, 2017 - link
You're joking.Dolphin doesn't run on my pixel c at anything resembling a useful frame rate (even when it actually works).
MajGenRelativity - Tuesday, March 14, 2017 - link
I'm not read up on all the lingo, but what does 16FFC stand for, and how does it differ from 16FF+?Ian Cutress - Tuesday, March 14, 2017 - link
Page 4:The Kirin 950 uses TSMC’s 16FF+ FinFET process, but HiSilicon switches to TSMC’s 16FFC FinFET process for the Kirin 960. The newer 16FFC process reduces manufacturing costs and die area to make it competitive in mid- to low-end markets, giving SoC vendors a migration path from 28nm. It also claims to reduce leakage and dynamic power by being able to run below 0.6V, making it suitable for wearable devices and IoT applications. Devices targeting price-sensitive markets, along with ultra low-power wearable devices, tend to run at lower frequencies, however, not 2.36GHz like Kirin 960. It’s possible that pushing the less performance-oriented 16FFC process, which targets lower voltages/frequencies, to higher frequencies that lay beyond its peak efficiency point may partially explain the higher power consumption relative to 16FF+.