Intel Core i9 9900K CofeeLake-R Review & Benchmarks – 2-channel DDR4 Cache & Memory Performance

What is “CofeeLake-R” CFL-R?

It is the “refresh” (updated) version of the 8th generation Intel Core architecture (CFL) – itself a minor stepping of the previous 7th generation “KabyLake” (KBL), itself a minor update of the 6th generation “SkyLake” (SKL). While ordinarily this would not be much of an event – this time we do have more significant changes:

• Patched vulnerabilities in hardware: this can help restore I/O workload performance degradation due to OS mitigations
  • Kernel Page Table Isolation (KPTI) aka “Meltdown” – Patched in hardware
  • L1TF/Foreshadow – Patched in hardware
  • (IBPB/IBRS) “Spectre 2” – OS mitigation needed
  • Speculative Store Bypass disabling (SSBD) “Spectre 4” – OS mitigation needed
• Increased core counts yet again: CFL-R top-end now has 8 cores, not 6.

Intel CPUs bore the brunt of the vulnerabilities disclosed at the start of 2018 with “Meltdown” operating system mitigations (KVA) likely having the biggest performance impact in I/O workloads. While modern features (e.g. PCID (process context id) acceleration) could help reduce performance impact somewhat on recent architectures (4th gen and newer) the impact can still be significant. The CFL-R hardware fixes (thus not needing KVA) may thus prove very important.

On the desktop we also see increased cores (again!) now up to 8 (thus 16 threads with HyperThreading) – double what KBL and SKL brought and matching AMD.

We also see increased clocks, mainly Turbo, but this still allows 1 or 2 cores to boost clocks higher than CFL could and thus help workloads not massively threaded. This can improve responsiveness as single tasks can be run at top speed when there is little thread utilization.

While rated TDP has not changed, in practice we are likely to see increased “real” power consumption especially due to higher clocks – with Turbo pushing power consumption even higher – close to SKL/KBL-X.

In this article we test CPU Core performance; please see our other articles on:

• Intel Core i7 9900K CofeeLake-R Review & Benchmarks – 8-core/16-thread CPU Performance
• Intel UHD 630 (Core i7 8700K, i9 9900K) – GPGPU Performance
• Intel Core i7 8700K CofeeLake Review & Benchmarks – CPU 6-core/12-thread Performance

Hardware Specifications

We are comparing the top-of-the-range Gen 8 Core i7 (8700K) with previous generation (6700K) and competing architectures with a view to upgrading to a mid-range high performance design.

Native Performance

We are testing native arithmetic, SIMD and cryptography performance using the highest performing instruction sets (AVX2, AVX, etc.). CFL-R supports most modern instruction sets (AVX2, FMA3) but not the latest SKL/KBL-X AVX512 nor a few others like SHA HWA (Atom, Ryzen).

Results Interpretation: Higher values (GOPS, MB/s, etc.) mean better performance.

Environment: Windows 10 x64 (1807), latest drivers. 2MB “large pages” were enabled and in use. Turbo / Boost was enabled on all configurations.

Spectre / Meltdown Windows Mitigations: all were enabled as per default (BTI enabled, RDCL/KVA enabled, PCID enabled).

Native Benchmarks		Intel i9-9900K CofeeLake-R	Intel i7-8700K CofeeLake	AMD Ryzen2 2700X Pinnacle Ridge	Intel i9-7900X SkyLake-X	Comments
	Total Inter-Core Bandwidth – Best (GB/s)	70.7 [+28%]	52.5	55.3	86	CFL-R finally overtakes Ryzen2 in inter-core bandwidth with almost 30% more bandwidth.
	Total Inter-Core Bandwidth – Worst (GB/s)	15.4 [-1%]	15.5	6.35	25.7	In worst-case pairs on Ryzen2 must go across CCXes – unlike Intel’s CPUs – thus CFL can muster over 2x more bandwidth in this case.
CFL-R manages good bandwidth improvement with its 2  extra cores allowing it to dominate Ryzen  2; worst-case bandwidth does not improve as the inter-core connector has remained the same
	Inter-Unit Latency – Same Core (ns)	13.4 [-7%]	14.4	13.5	15	With its faster clock, CFL-R manages lower inter-core latency with 7% drop.
	Inter-Unit Latency – Same Compute Unit (ns)	43.7 [-3%]	45	40	75	Within the same unit, Ryzen2 is again faster than CFL/R.
	Inter-Unit Latency – Different Compute Unit (ns)	–	–	115	–	Obviously going across CCXes is slow, about 3x slower which needs careful thread scheduling.
The multiple CCX designof Ryzen 2 still presents some challenges to programmers requiring threads to be carefully scheduled – thus the unified CFL-R just like CFL before it enjoys lower latencies throughout.
	Aggregated L1D Bandwidth (GB/s)	1890 [+39%]	1630	854	2220	Intel’s wide L1D in CFL/R means almost 2x more bandwidth than Ryzen 2.
	Aggregated L2 Bandwidth (GB/s)	618 [+8%]	571	720	985	But Ryzen2’s L2 caches are not only twice as big but also very wide – CFL/R surprisingly cannot beat it.
	Aggregated L3 Bandwidth (GB/s)	326 [=]	327	339	464	Ryzen’s 2 L3 caches also provide good bandwidth matching CFL’s unified L3 cache.
	Aggregated Memory (GB/s)	35.5 [=]	35.6	32.2	70	Running at 3200Mt’s obviously CFL enjoys higher bandwidth than Ryzen2 at 2667Mt’s but somehow the latter has better efficiency.
Nothing much has changed in CFL/R vs. old SKL/KBL thus while L1 caches are wide and thus fast – the L2, L3 are not as impressive and the memory controller while competitive it does not seem as efficient as Ryzen2 but is more stable at high data rates allowing for higher bandwidth.
	Data In-Page Random Latency (ns)	17.5 (3-10-21)	17.4 (4-11-20) [-73%]	63.4 (4-12-31)	25.5 (4-13-30)	While clock latencies have not changed w.s. old KBL/SKL, CFLR enjoys lower latencies due to higher data rates. Ryzen2 has problems here.
	Data Full Random Latency (ns)	54.3 (3-10-36)	53.4 (4-11-42) [-30%]	76.2 (4-12-32)	74 (4-13-62)	Out-of-page clock latencies have increased but still overall lower. Ryzen2 has almost caught up here.
	Data Sequential Latency (ns)	3.8 (3-10-11)	3.8 (4-11-12)	3.3 (4-6-7)	5.3 (4-12-12)	With sequential access, Ryzen2 is now faster as CFL/R’s clock latencies have not changed.
CFL-R does not improve over CFL (same memory controller) is lucky here as even Ryzen2 still has high latencies in random accesses (either in-page or full range) but manages to be faster with sequential access. Intel will need to improve going forward as clock latencies while good have really not improved at all.
	Code In-Page Random Latency (ns)	8.6 (2-9-19)	8.7 (2-10-21)	13.8 (4-9-24)	11.8 (4-14-25)	Code clock latencies also have not changed and again and while Ryzen2 performs a lot better, CFL/R manage to be ~35% faster.
	Code Full Random Latency (ns)	60.1 (2-9-48)	59.8 (2-10-48)	85.7 (4-14-49)	83.6 (4-15-74)	Out-of-page clock latencies also have not changed and here CFL/R is 20% faster over Ryzen2.
	Code Sequential Latency (ns)	4.3 (2-3-8)	4.5 (2-4-10)	7.4 (4-12-20)	6.8 (4-7-11)	Ryzen2 is competitive but again CFL/R manages to be almost 40% faster.
CFL/R does not improve over CFL but still dominates here and enjoys 30-40% less latency over Ryzen2 but the latter has improved a lot in time.
	Memory Update Transactional (MTPS)	73.3 [+36%]	54	5	59	Finally all top-end Intel CPUs have HLE enabled and working and thus enjoy huge performance increase.
	Memory Update Record Only (MTPS)	53.4 [+41%]	38	4.58	59	Nothing much changes here. CFL-R can do over 40% more transactions.

CFL-R does not really perform any different cache/memory wise vs. old CFL as the caches and memory controller are unchanged.

SiSoftware Official Ranker Scores

• Intel Core i9 9900K
• Intel Core i7 9700K

Final Thoughts / Conclusions

CFL-R just adds more cores, thus enjoys higher aggregated L1D/L2 bandiwdths vs CFL but the L3 is still disappointing – especially as now it has to feed 33% more cores/threads (8/16 vs 6/12). Latencies (in clocks) do not change either but as it can clock higher they do decrease in real terms (ns).

The memory controller is the very same (even running same firmware) thus performs the same though now it has to feed 33% more cores/threads (8/16 vs 6/12) thus when all cores/threads are used the aggregated bandwidth falls due to extra contention. In fairness Ryzen2 has the same issue (too many cores/threads for too little bandwidth) thus SKL/KBL-X is where you should be looking for more bandwidth.