| | | | | | |

A9 9420 vs Xeon Silver 4110


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

Using the multithread performance as a reference, the 9420 gets a score of 46.3 k points while the 4110 gets 281.5 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
670f00
50654
Core
Stoney Ridge
Skylake-SP
Architecture
Base frecuency
3 GHz
2.1 GHz
Boost frecuency
3.6 GHz
3 GHz
Socket
Micro-BGA
LGA3647
Cores/Threads
2/2
8/16
TDP
15 W
85 W
Cache L1 (d+i)
96+2x32 kB
8x32+8x32 kB
Cache L2
1024 kB
8x1024 kB
Cache L3
0 kB
11264 kB
Date
April 2017
July 2017
Mean monothread perf.
33.09k points
45.82k points
Mean multithread perf.
46.33k points
281.48k 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
9420
4110
Test#1 (Integers)
2.56k
3k (x1.17)
Test#2 (FP)
9.97k
12.29k (x1.23)
Test#3 (Generic, ZIP)
2.96k
3.64k (x1.23)
Test#1 (Memory)
2.12k
6.55k (x3.08)
TOTAL
17.61k
25.46k (x1.45)

Multithread

9420

4110
Test#1 (Integers)
3.88k
20.42k (x5.26)
Test#2 (FP)
9.26k
95.83k (x10.35)
Test#3 (Generic, ZIP)
3.9k
30.88k (x7.91)
Test#1 (Memory)
1.74k
9.7k (x5.57)
TOTAL
18.79k
156.83k (x8.35)

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
9420
4110
Test#1 (Integers)
11.74k
18.78k (x1.6)
Test#2 (FP)
16.05k
16.25k (x1.01)
Test#3 (Generic, ZIP)
3.26k
3.99k (x1.23)
Test#1 (Memory)
2.05k
6.8k (x3.32)
TOTAL
33.09k
45.82k (x1.38)

Multithread

9420

4110
Test#1 (Integers)
18.57k
133.55k (x7.19)
Test#2 (FP)
20.11k
108.43k (x5.39)
Test#3 (Generic, ZIP)
5.27k
30.02k (x5.69)
Test#1 (Memory)
2.38k
9.48k (x3.99)
TOTAL
46.33k
281.48k (x6.07)

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

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