| | | | | | |

Core i3-1005G1 vs i5-3470


Description
The i3-1005G1 is based on Ice Lake architecture while the i5-3470 is based on Ivy Bridge.

Using the multithread performance as a reference, the i3-1005G1 gets a score of 98.9 k points while the i5-3470 gets 94.1 k points.

Summarizing, the i3-1005G1 is 1.1 times faster than the i5-3470 . To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
706e5
306a9
Core
Ice Lake-U
Ivy Bridge
Architecture
Base frecuency
1.2 GHz
3.2 GHz
Boost frecuency
3.4 GHz
3.6 GHz
Socket
BGA 1526
LGA 1155
Cores/Threads
2/4
4/4
TDP
15 W
77 W
Cache L1 (d+i)
2x32+2x48 kB
4x32+4x32 kB
Cache L2
2x512 kB
4x256 kB
Cache L3
4096 kB
6144 kB
Date
August 2019
June 2012
Mean monothread perf.
50.62k points
31.71k points
Mean multithread perf.
116.63k points
94.15k 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
i3-1005G1
i5-3470
Test#1 (Integers)
3.07k
3.62k (x1.18)
Test#2 (FP)
13.76k
10.08k (x0.73)
Test#3 (Generic, ZIP)
8k
4.47k (x0.56)
Test#1 (Memory)
6.59k
5.39k (x0.82)
TOTAL
31.42k
23.56k (x0.75)

Multithread

i3-1005G1

i5-3470
Test#1 (Integers)
8k
13.66k (x1.71)
Test#2 (FP)
35.53k
37.73k (x1.06)
Test#3 (Generic, ZIP)
18.53k
17.07k (x0.92)
Test#1 (Memory)
4.3k
8.57k (x2)
TOTAL
66.36k
77.04k (x1.16)

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
i3-1005G1
i5-3470
Test#1 (Integers)
10.52k
12.26k (x1.17)
Test#2 (FP)
17.6k
11k (x0.62)
Test#3 (Generic, ZIP)
8.76k
4.76k (x0.54)
Test#1 (Memory)
6.55k
5.39k (x0.82)
TOTAL
43.44k
33.42k (x0.77)

Multithread

i3-1005G1

i5-3470
Test#1 (Integers)
26.86k
46.54k (x1.73)
Test#2 (FP)
42.38k
41.4k (x0.98)
Test#3 (Generic, ZIP)
20.65k
18.09k (x0.88)
Test#1 (Memory)
4.21k
8.69k (x2.06)
TOTAL
94.1k
114.73k (x1.22)

AVX optimized benchmark
The benchmark in mode II (AVX) is optimized to used 256 bits registers beside the first version of the Advanced Vector Extensions (AVX). The first AVX compatible CPU was released in 2011.
Monothread
i3-1005G1
i5-3470
Test#1 (Integers)
11.93k
11.51k (x0.96)
Test#2 (FP)
16.99k
11.12k (x0.65)
Test#3 (Generic, ZIP)
6.43k
4.44k (x0.69)
Test#1 (Memory)
4.59k
4.64k (x1.01)
TOTAL
39.93k
31.71k (x0.79)

Multithread

i3-1005G1

i5-3470
Test#1 (Integers)
29.07k
37.93k (x1.3)
Test#2 (FP)
46.14k
36.37k (x0.79)
Test#3 (Generic, ZIP)
19.47k
14.14k (x0.73)
Test#1 (Memory)
4.18k
5.71k (x1.37)
TOTAL
98.86k
94.15k (x0.95)

Performance/W
i3-1005G1
i5-3470
Test#1 (Integers)
1938 points/W
493 points/W
Test#2 (FP)
3076 points/W
472 points/W
Test#3 (Generic, ZIP)
1298 points/W
184 points/W
Test#1 (Memory)
279 points/W
74 points/W
TOTAL
6591 points/W
1223 points/W

Performance/GHz
i3-1005G1
i5-3470
Test#1 (Integers)
3510 points/GHz
3196 points/GHz
Test#2 (FP)
4996 points/GHz
3090 points/GHz
Test#3 (Generic, ZIP)
1890 points/GHz
1234 points/GHz
Test#1 (Memory)
1349 points/GHz
1289 points/GHz
TOTAL
11745 points/GHz
8809 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