CPU Performance: Memory and Power

Memory

With the same memory setup as the standard Skylake-style core, without any other improvements, we should expect Cannon Lake to perform exactly similar. Here’s a recap of the memory configuration:

Memory Comparison
  Skylake Desktop Cannon Lake
L1-D
Cache
32 KiB/core
8-way
32 KiB/core
8-way
L1-I
Cache
32 KiB/core
8-way
32 KiB/core
8-way
L2
Cache
256 KiB/core
4-way
256 KiB/core
4-way
L3
Cache
2 MiB/core
16-way
2 MiB/core
16-way
L3 Cache Type Inclusive Inclusive

For this test we put our Cannon Lake and Kaby Lake processors through our Memory Latency Checker test. Both systems had turbo modes disabled, forcing them to run at 2.2 GHz for parity and a direct microarchitecture comparison.

In this instance we see very little different between the two through the caches of the chip. Up to about 2MiB, both chips perform practically identical. Beyond this however, going out to main memory, there is a big discrepancy between the Kaby Lake and the Cannon Lake processor. In essence, accessing Cannon Lake main memory has an additional 50% latency.

At first we thought this was a bug. For both systems to have dual channel memory and running DDR4-2400, something had to be wrong. We double checked the setups – both systems were running in dual channel mode, giving the same memory bandwidth. The Cannon Lake processor was running at DDR4-2400 17-17-17, whereas the Kaby Lake system was at DDR4-2400 16-16-16 (due to memory SPD differences), which isn’t a big enough change to have such a big difference. The only reason we can come up with is that the memory controller on Cannon Lake must have additional overhead from the core to the memory controller – either a slower than expected PLL or something.

Power

Measuring the power of these thermally limited systems is somewhat frustrating in that we are at the whims of the quality of silicon at play. Bad silicon can get hotter faster, causing thermal throttling, or it depends on how the systems are set up for the peak power numbers. We use AIDA’s power monitoring tool to give us the power numbers for the CPUs during our POV-Ray test.

If you saw our How Intel Measures TDP article, you may be familiar with the concept of Power Limits. Intel processors have several power limits set in firmware – these values are set by the system manufacturer, and Intel provides ‘guidelines’. The two key values are Power Limit 1 (PL1), which describes a steady state scenario, and Power Limit 2 (PL2) which describes a turbo scenario. The reason why the system manufacturer gets control of these is because they might put a high wattage processor into a small form factor system, and need to manage performance – this is why we sometimes see a Core i7 be outperformed by a Core i5 in the same system design.

In most cases, PL1 is set to the TDP of the processor – in this case, it should be 15W. As shown on the graph above, this is true for our Core i3-8130U system, which shows a steady state power consumption of 15.0W. The Cannon Lake processor however only peaks at 12.6W.

PL2, the turbo power of the processor, can also be determined. For the Kaby Lake processor, we see a peak of 24.2 W, while the Cannon Lake has a turbo only of 18.7 W. We see that the turbo time period for the Cannon Lake processor is also much shorter, leading to a PL1 / steady state power much sooner.

Part of this is because we are pitting the Cannon Lake laptop against the Kaby Lake mini-PC. That being said, the Cannon Lake laptop is a beefy 15.6-inch unit, and has cooling for both the CPU and the discrete RX540 graphics, which isn’t doing much during this test except showing a basic 2D display.

That is arguably a lot more cooling than the mini-PC provides, plus we are cooling it with additional fans. These settings are more a function of both Lenovo doing what it wants to in firmware that we can’t change, but also the chip as engineered.

We can also calculate some level of efficiency here. By taking the area under the graph and correcting for the base line power (2.9 W on KBL, 3.6 W on CNL), we can calculate the total power consumed during the test. We get the following:

  • Core i3-8121U (CNL) consumes 867 mWh
  • Core i3-8130U (KBL) consumes 768 mWh

So this means that Cannon Lake is slower in AVX2 frequency, consumes more power, and scores 25% less in POV-Ray (see previous pages). That’s damning for the design.

To further look into this data, we disabled the turbo modes on both processors and ran our POV-Ray workload adjusting the number of threads being loaded and then measured the power consumption. In this scenario we are running at the same frequency, measuring the steady state power, and are thus at the whims of both the voltage set on both processors and the efficiency. Our tool also lets us measure the power of just the cores, so we’ll plot core power against threads loaded.

If this graph was the definitive graph, it shows that Cannon Lake is vastly inefficient (at 2.2 GHz) compared to Kaby Lake.

Our Testing Suite for 2018 and 2019 CPU Performance: SPEC2006 at 2.2 GHz
POST A COMMENT

129 Comments

View All Comments

  • Gondalf - Friday, January 25, 2019 - link

    For now they have nothing out in cpu departement, so i don't see any AMD bright year in front of us.
    I remember you we are already in 2019.
    Reply
  • vegajf51 - Friday, January 25, 2019 - link

    Icelake Desktop 3q 2020, intel will have another 14nm refresh before then. Reply
  • HStewart - Saturday, January 26, 2019 - link

    Intel is expected to release 10nm+ with Covey Lake by Christmas seasons. This canon lake chip is just a test chip. Reply
  • pugster - Friday, January 25, 2019 - link

    Thanks for the review. While the performance is not great, what about the power consumption compared with the 8130U? Reply
  • Yorgos - Friday, January 25, 2019 - link

    it's not great obviously when you are stuck at 2.2GHz, while the prev gen cpu with the same capabilities(except the avx) can go up to 3.4GHz.
    I bet the 8130 would've been faster even if configured at 10Watt TDP.
    Reply
  • Yorgos - Friday, January 25, 2019 - link

    ...and before jumping on me about that "stuck at 2.2GHz" let me report this:
    in certain loads the locked freq is slower than the unlocked one.
    What does this mean? it most probably means that the unlocked freq makes the cpu run hot, throttle and then try to balance between temperature and consumption.

    and a subnote on this. I think Intel should stop pushing the AVX instructions. It doesn't work as intended, it's not needed in most cases, especially when you have to design 256bit buses for 512bit data transfer on a low power cpu. Also it takes a lot of space on the die, it taxes the cache buses and it's useless when you disable your igpu(which is a good SIMD machine but not hUMA) and you have a dGPU up all the time just rendering your desktop.
    They should try focusing on HSA/hUMA on their cpus+igpus instead of integrating wide SIMD instructions inside their cores.
    Reply
  • 0ldman79 - Saturday, January 26, 2019 - link

    Thing is when AVX2 and AVX512 are used the performance increase can be rather massive.

    PCSX2, PS2 emulator, runs identically between my 3.9GHz Ivy Bridge Xeon (AVX) and my 2.8GHz i5 Skylake mobile (AVX2).

    AVX2 makes several games playable. You can choose your plugin and the AVX plugin cannot play Gran Turismo 4 @ 2.8GHz, the AVX2 plugin can.

    You may not find it useful, others do.
    Reply
  • HStewart - Saturday, January 26, 2019 - link

    It would be interesting to see the emulator re-factor to work with AVX 512 - it would like be twice the speed of AVX 2 Reply
  • levizx - Sunday, January 27, 2019 - link

    Nope, even with the simplest data set where AVX512 can perform twice the speed of AVX2 per cycle, the frequency has to drop significantly (~30% on Xeon Gold 5120 for example), so the upper limit is more like 40% gain. And that's PURE AVX512 code, you won't get that in real life. Assuming 50% AVX2 and 50% AVX512 code - that's a very generous assumption for non-datacentre usage, you'll have a 5% net gain. Reply
  • levizx - Sunday, January 27, 2019 - link

    5%~20% net gain, depending on how the scaling works. Reply

Log in

Don't have an account? Sign up now