Xeon E5-2667 v3 vs E5-1660 v3
Description
Both models E5-2667 v3 and E5-1660 v3 are based on Haswell architecture.
"Haswell is known as fourth-generation core too. It uses 22nm Tri-Gate FinFET (non-planar, ""3D"") technology. It includes four arithmetic logic unit (ALU) and three address generation unit (AGU). This generation is able to run Advanced Vector Extensions 2 (AVX2) instruction set among BMI1, BMI2, ABM and FMA3. At this time, Intel used new sockets: LGA 1150 for desktops and BGA1364 for the mobile CPUs."
Using the multithread performance as a reference, the E5-2667 v3 gets a score of 343.6 k points while the E5-1660 v3 gets 353 k points.
Summarizing, the E5-1660 v3 is 1 times faster than the E5-2667 v3. To get a proper comparison between both models, take a look to the data shown below.
"Haswell is known as fourth-generation core too. It uses 22nm Tri-Gate FinFET (non-planar, ""3D"") technology. It includes four arithmetic logic unit (ALU) and three address generation unit (AGU). This generation is able to run Advanced Vector Extensions 2 (AVX2) instruction set among BMI1, BMI2, ABM and FMA3. At this time, Intel used new sockets: LGA 1150 for desktops and BGA1364 for the mobile CPUs."
Using the multithread performance as a reference, the E5-2667 v3 gets a score of 343.6 k points while the E5-1660 v3 gets 353 k points.
Summarizing, the E5-1660 v3 is 1 times faster than the E5-2667 v3. To get a proper comparison between both models, take a look to the data shown below.
Specs
CPUID
306f2
306f2
Core
Haswell-EP
Haswell-EP
Base frecuency
3.2 GHz
3 GHz
Boost frecuency
3.6 GHz
3.5 GHz
Socket
LGA 2011-3
LGA 2011-3
Cores/Threads
8/16
8/16
TDP
135 W
140 W
Cache L1 (d+i)
8x32+8x32 kB
8x32+8x32 kB
Cache L2
8x256 kB
8x256 kB
Cache L3
20480 kB
20480 kB
Date
September 2014
September 2014
Mean monothread perf.
41.36k points
41.31k points
Mean multithread perf.
343.64k points
352.96k points
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
E5-2667 v3
E5-1660 v3
Test#1 (Integers)
20.99k
20.92k (x1)
Test#2 (FP)
11.9k
11.87k (x1)
Test#3 (Generic, ZIP)
4.72k
4.67k (x0.99)
Test#1 (Memory)
3.75k
3.85k (x1.03)
TOTAL
41.36k
41.31k (x1)
Multithread
E5-2667 v3
E5-1660 v3
Test#1 (Integers)
172.62k
174.59k (x1.01)
Test#2 (FP)
110.03k
116.47k (x1.06)
Test#3 (Generic, ZIP)
44.06k
44.79k (x1.02)
Test#1 (Memory)
16.93k
17.1k (x1.01)
TOTAL
343.64k
352.96k (x1.03)
Performance/W
E5-2667 v3
E5-1660 v3
Test#1 (Integers)
1279 points/W
1247 points/W
Test#2 (FP)
815 points/W
832 points/W
Test#3 (Generic, ZIP)
326 points/W
320 points/W
Test#1 (Memory)
125 points/W
122 points/W
TOTAL
2545 points/W
2521 points/W
Performance/GHz
E5-2667 v3
E5-1660 v3
Test#1 (Integers)
5830 points/GHz
5977 points/GHz
Test#2 (FP)
3305 points/GHz
3391 points/GHz
Test#3 (Generic, ZIP)
1311 points/GHz
1333 points/GHz
Test#1 (Memory)
1043 points/GHz
1101 points/GHz
TOTAL
11489 points/GHz
11802 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.
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]
Test#2 (FP) [points vs time]
Test#3 (Generic, ZIP) [points vs time]
Test#1 (Memory) [points vs time]
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.
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]
Test#2 (FP) [points vs time]
Test#3 (Generic, ZIP) [points vs time]
Test#1 (Memory) [points vs time]
Hardlimit Benchmark Central - Ver. 3.11.4