The Snapdragon 865 SoC: Beating Expectations

We’ve covered the Snapdragon 865 extensively over the last few months, and more recently did a performance preview of the chip on the Galaxy S20 Ultra:

It’s safe to say that Qualcomm managed to beat our expectations in terms of power efficiency improvements. Which is something we’ll go over in more detail in this piece as well.

At the heart of the Snapdragon 865 we find Arm’s newest Cortex-A77 CPU cores. The new microarchitecture is said to bring a 20-25% IPC improvement over its predecessors, and that’s where the new SoC derives most of its performance improvements from, as the clock frequencies of the cores are identical to that of last year’s Snapdragon 855.

Snapdragon 865 CPU Topology

One aspect where Qualcomm did improve the design is in doubling the shared L3 cache of the CPU cluster, going from 2MB to 4MB. Not only does this further improve the performance of the CPUs by allowing for more data to be cached on-chip, but Qualcomm has explained that one of the primary reasons for this was to also improve power efficiency of the SoC by reducing how often the SoC has to access the DRAM, which is a relatively power-expensive operation.

The chip still has a 3MB system level cache that serves the various IP blocks on the SoC – it’s again meant to not only improve performance but also improve power efficiency as it avoids external memory accesses. The memory subsystem here is smart and detects when to bypass this cache when there’s latency-sensitive workloads, and in general we’ll see some massive memory subsystem improvements on the part of the Snapdragon 865 in a later dedicated section.

All the CPUs being in the same cluster and cache hierarchy means that the core-to-core latencies are relatively uniform, only differing based on their frequencies and lower level cache access latencies. It’s not too much of an exciting metric here, but it’s important context to have as we’ll consider the Exynos 990’s CPU topology in just a bit.

Again, we’ve covered the Snapdragon 865 quite extensively in the above linked articles so I recommend reading them again for other details on other parts of the new chip, such as the new ISP, DSP, and GPU details. However, one aspect that’s very defining for the flagship Qualcomm chipset this year is that the company is separating the modem from the SoC – essentially making the SoC just an application processor for this generation.

The external nature of the X55 modem has a few implications: first of all, there’s an additional component on the motherboard which vendors will have to make space for, which means additional cost. Secondly, there’s the big question of how power efficiency will be affected by the external modem. We’ve seen Apple devices perform excellently over the years while never having an integrated modem, and I feel like the Snapdragon 865 and X55 also fall into this classification, as I haven’t seen any major differences in efficiency due to the external nature of the modem.

Design, Continued: An Ultra Mega Phone The Exynos 990 SoC: Last of Custom CPUs
Comments Locked


View All Comments

  • toyeboy89 - Friday, April 3, 2020 - link

    I'm really amazed in the fact that the iPhone XR is still beating snapdragon 865 in GFXBench in both peak and sustained performance. I am hoping the OnePlus 8 has better sustained performance.
  • TMCThomas - Friday, April 3, 2020 - link

    Amazing review! Always wait for this one before getting a new samsung. And I won't be getting any of the s20 phones. For me the kind of feel like "beta" phones. The 120hz which is not quite ready for 1440p yet, the underutilized 108mp camera, the space zoom which is blurry, the camera hole still being there the big camera bump and so on. I think all these features and more could be way more refined with the next galaxy s which I'll be waiting for. Also the poor exynos 990 performance especially the GPU part is just unacceptable to me. Especially with it probably being a lot better next year, so I'll skip this year
  • wheeliebin - Friday, April 3, 2020 - link

    Thanks Andrei, really good review!

    I have read many users complain about extra crazy post-processing on the S10/S20 series when there is a face detected in the frame. i.e. the phone will apply an aggressive 'smooth skin' filter that you can't disable unless you shoot RAW. I was hoping that your review might touch on this however there were no people in your example shots so perhaps you didn't get a chance to experience the problem. I wonder if you have heard of this issue and can replicate it yourself with the S20 range?
  • anonomouse - Friday, April 3, 2020 - link

    Hi Andrei, did you also run the bandwidth and MLP sweeps from previous reviews? Last year you noted the Snapdragon 855/A76 had peculiar behavior in the L1, and it would be also interesting to see if there are any MLP changes in both the SD865 and the Exynos.
  • anonomouse - Friday, April 3, 2020 - link

    Also, any idea why the new scores for these in 403.gcc seems to be worse than their previous generation products? In particular the score for the SD865 in these S20s is substantially worse than the SD865 score from the QRD preview article.
  • Andrei Frumusanu - Saturday, April 4, 2020 - link

    Yes I know. I don't know why that happens. I also got a V60 now and the scores there are higher, I'm wondering if there's something with Samsungs shared libraries.
  • anonomouse - Sunday, April 5, 2020 - link

    What type of compile flags are used for these binaries? Are they the same for all of the tested binaries (or even same binary on each given platform)? Are LTO or PGO used (and if not why not)?

    I'm also not convinced of this statement from the article:

    "I had mentioned that the 7LPP process is quite a wildcard in the comparisons here. Luckily, I’ve been able to get my hands on a Snapdragon 765G, another SoC that’s manufactured on Samsung’s EUV process. It’s also quite a nice comparison as we’re able to compare that chip’s performance A76 cores at 2.4GHz to the middle A76 cores of the Exynos 990 which run at 2.5GHz. Performance and power between the two chips here pretty much match each other, and a clearly worse than other TSMC A76-based SoCs, especially the Kirin 990’s. The only conclusion here is that Samsung’s 7LPP node is quite behind TSMC’s N7/N7P/N7+ nodes when it comes to power efficiency – anywhere from 20 to 30%."

    Both the energy consumed and the performance scores for both of these A76's seem to also very closely track the "mid" 2.43Ghz A76's on the TSMC-fabbed SD855 - all of which have similar L2's and similar frequencies, but possibly differ significantly (to the point of being suboptimal on latency) on the memory hierarchy and SoC beyond that - which greatly affects many of the SPEC workloads. All of these may also have implementation targets. Given this, is it really conclusive that the Samsung process is truly 20-30% worse in energy efficiency? Granted, things will probably not look pretty next year when TSMC is on a true 5nm and Samsung is not.
  • Andrei Frumusanu - Monday, April 6, 2020 - link

    The test is just -Ofast without any other addition. LTO wasn't/isn't in a good state on the Android NDK - it's something to look into in maybe a new binary revision.

    As for the 855 figures, well, that's also on an earlier 7nm. HiSilicon did a lot better in terms of they physical implementation. If not against N7, 7LPP clearly has a disadvantage against N7P/N7+.
  • Andrei Frumusanu - Saturday, April 4, 2020 - link

    I'll add them in, that test takes a whole day and I needed the phones doing battery tests and other stuff.
  • dad_at - Saturday, April 4, 2020 - link

    Again, your S10+ Exynos results in pc mark are false as of 2020. In performance mode I easily get 9500 work 2.0 overall, about 9600 in browser bench, 21K in photo editing. PC mark in general is inconsistent, irrelevant benchmark, not representative of actual performance in daily usage. The same about these ancient SPEC synthetics. No one uses these for performance evaluation now.

Log in

Don't have an account? Sign up now