| | | | | | |

Ryzen 9 3900 vs Core i9-10980XE


Description
The 3900 is based on Zen 2 architecture while the i9-10980XE is based on Cascade Lake.

Using the multithread performance as a reference, the 3900 gets a score of 687.5 k points while the i9-10980XE gets 1241.1 k points.

Summarizing, the i9-10980XE is 1.8 times faster than the 3900. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
870f10
50657
Core
Matisse
Cascade Lake-X
Architecture
Base frecuency
3.1 GHz
3 GHz
Boost frecuency
4.3 GHz
4.8 GHz
Socket
AM4
LGA 2066
Cores/Threads
12/24
18/36
TDP
65 W
165 W
Cache L1 (d+i)
12x32+12x32 kB
18x32+18x32 kB
Cache L2
12x512 kB
18x1024 kB
Cache L3
4x16384 kB
25344 kB
Date
September 2019
November 2019
Mean monothread perf.
74.97k points
68.33k points
Mean multithread perf.
687.5k points
1241.12k 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
3900
i9-10980XE
Test#1 (Integers)
4.39k
4.52k (x1.03)
Test#2 (FP)
16.99k
17.94k (x1.06)
Test#3 (Generic, ZIP)
7.74k
5.5k (x0.71)
Test#1 (Memory)
24.57k
9.64k (x0.39)
TOTAL
53.69k
37.6k (x0.7)

Multithread

3900

i9-10980XE
Test#1 (Integers)
48.45k
84.93k (x1.75)
Test#2 (FP)
230.81k
409.55k (x1.77)
Test#3 (Generic, ZIP)
122.82k
122.88k (x1)
Test#1 (Memory)
54.73k
22.12k (x0.4)
TOTAL
456.8k
639.48k (x1.4)

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
3900
i9-10980XE
Test#1 (Integers)
16.59k
16.03k (x0.97)
Test#2 (FP)
24.55k
23.95k (x0.98)
Test#3 (Generic, ZIP)
8.55k
6.03k (x0.71)
Test#1 (Memory)
26.28k
9.37k (x0.36)
TOTAL
75.97k
55.38k (x0.73)

Multithread

3900

i9-10980XE
Test#1 (Integers)
181.33k
316.11k (x1.74)
Test#2 (FP)
261.01k
503.75k (x1.93)
Test#3 (Generic, ZIP)
126.62k
131.82k (x1.04)
Test#1 (Memory)
66.64k
22.24k (x0.33)
TOTAL
635.6k
973.92k (x1.53)

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
3900
i9-10980XE
Test#1 (Integers)
16.16k
16.4k (x1.01)
Test#2 (FP)
23.75k
20.97k (x0.88)
Test#3 (Generic, ZIP)
9.32k
5.82k (x0.62)
Test#1 (Memory)
23.17k
8.86k (x0.38)
TOTAL
72.4k
52.05k (x0.72)

Multithread

3900

i9-10980XE
Test#1 (Integers)
174.73k
316.84k (x1.81)
Test#2 (FP)
275.63k
461.09k (x1.67)
Test#3 (Generic, ZIP)
125.65k
129.82k (x1.03)
Test#1 (Memory)
44.17k
22.09k (x0.5)
TOTAL
620.18k
929.85k (x1.5)

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
3900
i9-10980XE
Test#1 (Integers)
16.85k
29.53k (x1.75)
Test#2 (FP)
26.03k
22.59k (x0.87)
Test#3 (Generic, ZIP)
9.54k
5.99k (x0.63)
Test#1 (Memory)
22.55k
10.22k (x0.45)
TOTAL
74.97k
68.33k (x0.91)

Multithread

3900

i9-10980XE
Test#1 (Integers)
229.41k
587.06k (x2.56)
Test#2 (FP)
292.81k
496.41k (x1.7)
Test#3 (Generic, ZIP)
128.48k
134.66k (x1.05)
Test#1 (Memory)
36.8k
22.99k (x0.62)
TOTAL
687.5k
1241.12k (x1.81)

Performance/W
3900
i9-10980XE
Test#1 (Integers)
3529 points/W
3558 points/W
Test#2 (FP)
4505 points/W
3009 points/W
Test#3 (Generic, ZIP)
1977 points/W
816 points/W
Test#1 (Memory)
566 points/W
139 points/W
TOTAL
10577 points/W
7522 points/W

Performance/GHz
3900
i9-10980XE
Test#1 (Integers)
3920 points/GHz
6153 points/GHz
Test#2 (FP)
6054 points/GHz
4707 points/GHz
Test#3 (Generic, ZIP)
2218 points/GHz
1248 points/GHz
Test#1 (Memory)
5243 points/GHz
2129 points/GHz
TOTAL
17435 points/GHz
14236 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