Final Thoughts

2018 has been a very successful year for Qualcomm and the Snapdragon 845. The company had provided an extremely solid and well-rounded SoC for device vendors to build their flagship devices on- and by the looks of it the Snapdragon 855 continues this trend.

I’ve been a bit sceptical about the merits of Qualcomm’s 1+3 CPU configuration, however after seeing the preliminary performance and power efficiency figures of the new prime core on the new chipset, I’m not nearly as concerned. We reserve any final verdict for when we will have tested final commercial devices, as that’s where in the end we’ll also see the efficiency effect of the non-prime cores, and how they’ll position themselves against the competition.

Performance wise, the Snapdragon 855 is a bit odd. In steady-state workloads like SPEC the chipset is seemingly performing very well and matches or exceeds the new Kirin 980. Here Qualcomm’s changes to the CPU microarchitecture might even actually be visible in the test results, which is a nice feat. Unfortunately the memory subsystem still seems to include some of DRAM latency regressions that we also saw in the Snapdragon 845, both which are due to Qualcomm’s system level cache.

Real-world performance, while still excellent, doesn’t quite manage to reach my expectations I had for the chipset. Here for whatever reason, the chip’s improvements are not nearly as pronounced as in the more synthetic tests. Again the odd thing is that the Kirin 980 still manages to beat the Snapdragon 855 in near most of these workloads. Qualcomm had reasoned that the microarchitectural changes to the CPU were meant to help web browser performance, yet it’s here where the chip slightly lags behind the competition – I do wonder if this is a case of the CPU again being limited by either Qualcomm’s choice of more conservative caches or due to the latency penalty of the system cache.

Although the performance shown today is exemplary, it still does look maybe a little rough around the edges in some of our system performance tests – here maybe Qualcomm will be able to investigate and further improve things until we actually see commercial devices.

Whether the system performance will be improved in final devices or not, what is clear though is that power efficiency looks outstanding. Qualcomm had me worried as the PR teams had avoided talking about efficiency during the chipset’s launch, but the results today (even if they’ll need to be verified), look very promising and should result in notable battery life improvements in 2019’s devices.

On the GPU side of things, Qualcomm’s more muted performance projections of 20% were because the company has again focused a good part of the process improvements into bringing the overall power back down from the usually higher levels that we saw on the Snapdragon 845.

Overall – the Snapdragon 855 looks to be another extremely well executed SoC from Qualcomm, and I’m looking forward to validating the results and testing out the first commercial devices once they become available.

GPU Performance - Returning To Lower Power
Comments Locked

132 Comments

View All Comments

  • Rudde - Friday, January 18, 2019 - link

    Don't cherry-pick results.
  • Rudde - Friday, January 18, 2019 - link

    What is raw performance? I could calculate som fused multiplies per second for you, but is that 'raw performance'?
  • HStewart - Tuesday, January 15, 2019 - link

    I differ with a lot on this - I think A12/A12X and other ARM related device actually perceive faster because of marketing. Also with App architexture of the OS running on such devices hide the actual performance of chips - I think it specifically depends on what you using the device. For normal word processing, emails and internet - it can easy be shown that same as U series - and this depends on which model - likely dual core x86 and possibly even AMD notebooks - but not a 4+ ghz laptop like my Dell XPS 15 2in1. Keep in mind on a phone and even android tablet or iPad there is less screen to drive. I am talking about real professional software and not apps

    One thing is interesting about 855 design is big core designed - running the primary core at higher speed then other 3 primary core - is smart - this means the primary thread is running at higher speed. I assume that smaller cores would be use threads for background tasks. Intel has a similar designed large single core and 3 minor atom based cores - I would think that device is closer on performance to A12 based devices not the U series.

    My big question is that I think it hard to actual compare performance between any x86 base and any ARM based. It depends on designed of OS and applications running on devices. I would be 100% sure any software that uses AVX 512 would blow any ARM based application with similar abilities. In fact with AVX 512 application it would be big difference with AVX 2 based computer.

    All I am saying is performance depends on application running, not just web browsing and other things
  • HStewart - Tuesday, January 15, 2019 - link

    One thing also - the speed of cpu, number of cores, or even node process does not make the performance of device - it how it used with architexture inside that makes the difference.
  • Wilco1 - Tuesday, January 15, 2019 - link

    You're absolutely wrong, A12X can keep up with your beloved laptop - this is the latest and fastest variant: http://browser.geekbench.com/v4/cpu/compare/109702...

    Single-threaded integer score is within 2.5%. Mind you that's a 10W SoC compared with a 65W CPU! I'm awaiting your list of excuses how that is possible...
  • goatfajitas - Tuesday, January 15, 2019 - link

    One test on one specific thing. Try 100's
  • Wilco1 - Tuesday, January 15, 2019 - link

    No it's not one test, nor one specific thing. Like most benchmarks there are many different tests and the average is reported.
  • goatfajitas - Tuesday, January 15, 2019 - link

    My point is that in a world of benchmarks, you are looking at it very myopically. ARM isnt anywhere near as fast as x86 in raw power. Very good and super efficient at alot of multimedia tasks though.
  • Wilco1 - Tuesday, January 15, 2019 - link

    What benchmarks??? Name another cross platform benchmark which is NOT a useless browser test. Apart from SPEC, Geekbench is one of the very few benchmarks that allow reasonable cross platform comparisons.
  • goatfajitas - Tuesday, January 15, 2019 - link

    "cross platform" benchmarks are virtually useless. Your grasp of benchmarking in general needs work. It's not apples to apples.

Log in

Don't have an account? Sign up now