| | | | | | |

Xeon E5-2697 v2 vs E5-2670 v3


Description
The E5-2697 v2 is based on Ivy Bridge architecture while the E5-2670 v3 is based on Haswell.

Using the multithread performance as a reference, the E5-2697 v2 gets a score of 633.1 k points while the E5-2670 v3 gets 338.7 k points.

Summarizing, the E5-2697 v2 is 1.9 times faster than the E5-2670 v3. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
306e4
306f2
Core
Ivy Bridge-EP
Haswell-EP
Architecture
Base frecuency
2.7 GHz
2.3 GHz
Boost frecuency
3.5 GHz
3.1 GHz
Socket
LGA 2011
LGA 2011-3
Cores/Threads
12 /24
12/24
TDP
130 W
120 W
Cache L1 (d+i)
12x32+12x32 kB
12x32+12x32 kB
Cache L2
12x256 kB
12x256 kB
Cache L3
30720 kB
30720 kB
Date
September 2013
September 2014
Mean monothread perf.
27.92k points
36.41k points
Mean multithread perf.
633.14k points
452.9k 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
E5-2697 v2
E5-2670 v3
Test#1 (Integers)
3.44k
3.52k (x1.02)
Test#2 (FP)
9.54k
8.68k (x0.91)
Test#3 (Generic, ZIP)
4.21k
2.83k (x0.67)
Test#1 (Memory)
3.94k
3.13k (x0.79)
TOTAL
21.13k
18.17k (x0.86)

Multithread

E5-2697 v2

E5-2670 v3
Test#1 (Integers)
73.17k
32.7k (x0.45)
Test#2 (FP)
230.83k
105.18k (x0.46)
Test#3 (Generic, ZIP)
107.14k
35.18k (x0.33)
Test#1 (Memory)
7.99k
9.54k (x1.19)
TOTAL
419.12k
182.59k (x0.44)

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
E5-2697 v2
E5-2670 v3
Test#1 (Integers)
10.64k
11.08k (x1.04)
Test#2 (FP)
9.53k
9.57k (x1)
Test#3 (Generic, ZIP)
4.12k
3.18k (x0.77)
Test#1 (Memory)
3.62k
3.28k (x0.91)
TOTAL
27.91k
27.11k (x0.97)

Multithread

E5-2697 v2

E5-2670 v3
Test#1 (Integers)
252.69k
139.07k (x0.55)
Test#2 (FP)
238.68k
133.47k (x0.56)
Test#3 (Generic, ZIP)
115.5k
46.6k (x0.4)
Test#1 (Memory)
10.98k
6.65k (x0.61)
TOTAL
617.84k
325.78k (x0.53)

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
E5-2697 v2
E5-2670 v3
Test#1 (Integers)
10.53k
11.18k (x1.06)
Test#2 (FP)
9.98k
10.06k (x1.01)
Test#3 (Generic, ZIP)
4k
3.21k (x0.8)
Test#1 (Memory)
3.4k
3.12k (x0.92)
TOTAL
27.92k
27.57k (x0.99)

Multithread

E5-2697 v2

E5-2670 v3
Test#1 (Integers)
247.05k
140.98k (x0.57)
Test#2 (FP)
264.43k
145.37k (x0.55)
Test#3 (Generic, ZIP)
111.01k
45.62k (x0.41)
Test#1 (Memory)
10.66k
6.75k (x0.63)
TOTAL
633.14k
338.73k (x0.54)

Performance/W
E5-2697 v2
E5-2670 v3
Test#1 (Integers)
1900 points/W
1175 points/W
Test#2 (FP)
2034 points/W
1211 points/W
Test#3 (Generic, ZIP)
854 points/W
380 points/W
Test#1 (Memory)
82 points/W
56 points/W
TOTAL
4870 points/W
2823 points/W

Performance/GHz
E5-2697 v2
E5-2670 v3
Test#1 (Integers)
3007 points/GHz
3607 points/GHz
Test#2 (FP)
2852 points/GHz
3245 points/GHz
Test#3 (Generic, ZIP)
1144 points/GHz
1035 points/GHz
Test#1 (Memory)
973 points/GHz
1007 points/GHz
TOTAL
7976 points/GHz
8894 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