| | | | | | |

Xeon Silver 4110 vs A9 9420


Description
The 4110 is based on Skylake architecture while the 9420 is based on Excavator.

Using the multithread performance as a reference, the 4110 gets a score of 281.5 k points while the 9420 gets 46.3 k points.

Summarizing, the 4110 is 6.1 times faster than the 9420. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
50654
670f00
Core
Skylake-SP
Stoney Ridge
Architecture
Base frecuency
2.1 GHz
3 GHz
Boost frecuency
3 GHz
3.6 GHz
Socket
LGA3647
Micro-BGA
Cores/Threads
8/16
2/2
TDP
85 W
15 W
Cache L1 (d+i)
8x32+8x32 kB
96+2x32 kB
Cache L2
8x1024 kB
1024 kB
Cache L3
11264 kB
0 kB
Date
July 2017
April 2017
Mean monothread perf.
45.82k points
33.09k points
Mean multithread perf.
281.48k points
46.33k points

Non-optimized benchmark
The benchmark in Mode 0 (FPU) measures cpu performance with non-optimized software. It uses the basic µinstructions from the i386 architecture with the i387 floating point unit. This mode is compatible with all CPUs so it's practical to compare very different CPUs
Monothread
4110
9420
Test#1 (Integers)
3k
2.56k (x0.86)
Test#2 (FP)
12.29k
9.97k (x0.81)
Test#3 (Generic, ZIP)
3.64k
2.96k (x0.81)
Test#1 (Memory)
6.55k
2.12k (x0.32)
TOTAL
25.46k
17.61k (x0.69)

Multithread

4110

9420
Test#1 (Integers)
20.42k
3.88k (x0.19)
Test#2 (FP)
95.83k
9.26k (x0.1)
Test#3 (Generic, ZIP)
30.88k
3.9k (x0.13)
Test#1 (Memory)
9.7k
1.74k (x0.18)
TOTAL
156.83k
18.79k (x0.12)

AVX2 optimized benchmark
The benchmark in mode III (AVX2), like AVX1, is optimized to used 256 bits registers beside the second version of the Advanced Vector Extensions (AVX). The first AVX2 compatible CPU was released in 2013.
Monothread
4110
9420
Test#1 (Integers)
18.78k
11.74k (x0.62)
Test#2 (FP)
16.25k
16.05k (x0.99)
Test#3 (Generic, ZIP)
3.99k
3.26k (x0.82)
Test#1 (Memory)
6.8k
2.05k (x0.3)
TOTAL
45.82k
33.09k (x0.72)

Multithread

4110

9420
Test#1 (Integers)
133.55k
18.57k (x0.14)
Test#2 (FP)
108.43k
20.11k (x0.19)
Test#3 (Generic, ZIP)
30.02k
5.27k (x0.18)
Test#1 (Memory)
9.48k
2.38k (x0.25)
TOTAL
281.48k
46.33k (x0.16)

Performance/W
4110
9420
Test#1 (Integers)
1571 points/W
1238 points/W
Test#2 (FP)
1276 points/W
1340 points/W
Test#3 (Generic, ZIP)
353 points/W
352 points/W
Test#1 (Memory)
112 points/W
159 points/W
TOTAL
3311 points/W
3089 points/W

Performance/GHz
4110
9420
Test#1 (Integers)
6259 points/GHz
3260 points/GHz
Test#2 (FP)
5417 points/GHz
4458 points/GHz
Test#3 (Generic, ZIP)
1330 points/GHz
905 points/GHz
Test#1 (Memory)
2267 points/GHz
570 points/GHz
TOTAL
15273 points/GHz
9192 points/GHz

Monothread performance graph
Monothread performance graphics gives the performance vs time. They are useful to measure the time it takes to the CPU to reach the maximum performance.

Usually, CPU's performance will be steady during these tests but if it has a slow frequency strategy, the first samples will show a lower score.


Test#1 (Integers) [points vs time]

grafica bm.hardlimit.com


Test#2 (FP) [points vs time]

grafica bm.hardlimit.com


Test#3 (Generic, ZIP) [points vs time]

grafica bm.hardlimit.com


Test#1 (Memory) [points vs time]

grafica bm.hardlimit.com

Multithread performance graph
Multithread graphs measure the performance against a heavy load during certain time.

If CPU's TDP doesn't limit the frequency and the machine is properly cooled, performance should remain steady vs time. Otherwise, the performance score will oscillate or decrease over time.


Test#1 (Integers) [points vs time]

grafica bm.hardlimit.com


Test#2 (FP) [points vs time]

grafica bm.hardlimit.com


Test#3 (Generic, ZIP) [points vs time]

grafica bm.hardlimit.com


Test#1 (Memory) [points vs time]

grafica bm.hardlimit.com

Hardlimit Benchmark Central - Ver. 3.11.4