| | | | | | |

Celeron J1900 vs G3900TE


Description
The J1900 is based on Silvermont architecture while the G3900TE is based on Skylake.

Using the multithread performance as a reference, the J1900 gets a score of 33.2 k points while the G3900TE gets 49.3 k points.

Summarizing, the G3900TE is 1.5 times faster than the J1900. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
30679
506e3
Core
Bay Trail-D
Skylake-S
Architecture
Base frecuency
2 GHz
2.3 GHz
Boost frecuency
2.42 GHz
2.3 GHz
Socket
BGA 1170
LGA 1151
Cores/Threads
4/4
2/2
TDP
10 W
35 W
Cache L1 (d+i)
4x32+4x24 kB
2x32+2x32 kB
Cache L2
2x1024 kB
2x256 kB
Cache L3
0 kB
2048 kB
Date
November 2013
December 2015
Mean monothread perf.
9.75k points
27.33k points
Mean multithread perf.
33.25k points
49.34k 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
J1900
G3900TE
Test#1 (Integers)
1.24k
2.36k (x1.9)
Test#2 (FP)
1.77k
9.79k (x5.54)
Test#3 (Generic, ZIP)
1.65k
3.08k (x1.86)
Test#1 (Memory)
1.15k
3.36k (x2.93)
TOTAL
5.81k
18.59k (x3.2)

Multithread

J1900

G3900TE
Test#1 (Integers)
4.68k
4.45k (x0.95)
Test#2 (FP)
6.67k
18.16k (x2.72)
Test#3 (Generic, ZIP)
6.15k
5.48k (x0.89)
Test#1 (Memory)
1.45k
3.5k (x2.41)
TOTAL
18.96k
31.59k (x1.67)

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
J1900
G3900TE
Test#1 (Integers)
4.58k
8.56k (x1.87)
Test#2 (FP)
2.32k
12.25k (x5.29)
Test#3 (Generic, ZIP)
1.72k
3.17k (x1.85)
Test#1 (Memory)
1.14k
3.35k (x2.95)
TOTAL
9.75k
27.33k (x2.8)

Multithread

J1900

G3900TE
Test#1 (Integers)
17.16k
16.42k (x0.96)
Test#2 (FP)
8.42k
23.61k (x2.8)
Test#3 (Generic, ZIP)
6.22k
5.75k (x0.92)
Test#1 (Memory)
1.44k
3.55k (x2.46)
TOTAL
33.25k
49.34k (x1.48)

Performance/W
J1900
G3900TE
Test#1 (Integers)
1716 points/W
469 points/W
Test#2 (FP)
842 points/W
675 points/W
Test#3 (Generic, ZIP)
622 points/W
164 points/W
Test#1 (Memory)
144 points/W
101 points/W
TOTAL
3325 points/W
1410 points/W

Performance/GHz
J1900
G3900TE
Test#1 (Integers)
1893 points/GHz
3724 points/GHz
Test#2 (FP)
957 points/GHz
5326 points/GHz
Test#3 (Generic, ZIP)
709 points/GHz
1380 points/GHz
Test#1 (Memory)
470 points/GHz
1455 points/GHz
TOTAL
4029 points/GHz
11885 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