| | | | | | |

A9 9420 vs Celeron 3855U


Description
The 9420 is based on Excavator architecture while the 3855U is based on Skylake.

Using the multithread performance as a reference, the 9420 gets a score of 28.9 k points while the 3855U gets 24.6 k points.

Summarizing, the 9420 is 1.2 times faster than the 3855U. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
670f00
406e3
Core
Stoney Ridge
Skylake-U
Architecture
Base frecuency
3 GHz
1.6 GHz
Boost frecuency
3.6 GHz
1.6 GHz
Socket
Micro-BGA
BGA1356
Cores/Threads
2/2
2/2
TDP
15 W
15 W
Cache L1 (d+i)
96+2x32 kB
2x32+2x32 kB
Cache L2
1024 kB
2x256 kB
Cache L3
0 kB
2096 kB
Date
April 2017
December 2015
Mean monothread perf.
33.09k points
19.24k points
Mean multithread perf.
46.33k points
24.56k 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
3855U
Test#1 (Integers)
2.56k
1.64k (x0.64)
Test#2 (FP)
9.97k
6.72k (x0.67)
Test#3 (Generic, ZIP)
2.96k
2.06k (x0.7)
Test#1 (Memory)
2.12k
2.4k (x1.13)
TOTAL
17.61k
12.83k (x0.73)

Multithread

9420

3855U
Test#1 (Integers)
3.88k
2.2k (x0.57)
Test#2 (FP)
9.26k
9.59k (x1.04)
Test#3 (Generic, ZIP)
3.9k
2.76k (x0.71)
Test#1 (Memory)
1.74k
2.27k (x1.3)
TOTAL
18.79k
16.81k (x0.89)

SSE3 optimized benchmark
The benchmark in mode I (SSE) is optimized for the use of SIMD instructions with 128 bits register and the SSE set up to version 3. Nearly every modern CPU has support for this mode.
Monothread
9420
3855U
Test#1 (Integers)
8.41k
5.93k (x0.71)
Test#2 (FP)
11.48k
8.53k (x0.74)
Test#3 (Generic, ZIP)
3.29k
2.28k (x0.69)
Test#1 (Memory)
2.24k
2.5k (x1.12)
TOTAL
25.41k
19.24k (x0.76)

Multithread

9420

3855U
Test#1 (Integers)
11.25k
7.87k (x0.7)
Test#2 (FP)
13.69k
11.28k (x0.82)
Test#3 (Generic, ZIP)
2.09k
2.91k (x1.4)
Test#1 (Memory)
1.87k
2.5k (x1.34)
TOTAL
28.89k
24.56k (x0.85)

Performance/W
9420
3855U
Test#1 (Integers)
750 points/W
524 points/W
Test#2 (FP)
912 points/W
752 points/W
Test#3 (Generic, ZIP)
139 points/W
194 points/W
Test#1 (Memory)
125 points/W
166 points/W
TOTAL
1926 points/W
1637 points/W

Performance/GHz
9420
3855U
Test#1 (Integers)
2336 points/GHz
3706 points/GHz
Test#2 (FP)
3188 points/GHz
5333 points/GHz
Test#3 (Generic, ZIP)
913 points/GHz
1423 points/GHz
Test#1 (Memory)
621 points/GHz
1562 points/GHz
TOTAL
7058 points/GHz
12024 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