| | | | | | |

Ryzen 9 3900 vs Core i9-13900K


Description
The 3900 is based on Zen 2 architecture while the i9-13900K is based on Raptor Lake.

Using the multithread performance as a reference, the 3900 gets a score of 687.5 k points while the i9-13900K gets 1763.1 k points.

Summarizing, the i9-13900K is 2.6 times faster than the 3900. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
870f10
b0671
Core
Matisse
Raptor Lake-S
Architecture
Zen 2
Base frecuency
3.1 GHz
2.2 GHz
Boost frecuency
4.3 GHz
5.8 GHz
Socket
AM4
LGA 1700
Cores/Threads
12/24
24/32
TDP
65 W
125 W
Cache L1 (d+i)
12x32+12x32 kB
8x32/16x64+8x48/16x32 kB
Cache L2
12x512 kB
8x2048+4x4096 kB
Cache L3
4x16384 kB
36864 kB
Date
September 2019
October 2022
Mean monothread perf.
74.97k points
129.15k points
Mean multithread perf.
687.5k points
1763.08k 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
i9-13900K
Test#1 (Integers)
4.39k
9.77k (x2.23)
Test#2 (FP)
16.99k
24.52k (x1.44)
Test#3 (Generic, ZIP)
7.74k
17.17k (x2.22)
Test#1 (Memory)
24.57k
21.37k (x0.87)
TOTAL
53.69k
72.83k (x1.36)

Multithread

3900

i9-13900K
Test#1 (Integers)
48.45k
145.21k (x3)
Test#2 (FP)
230.81k
444.95k (x1.93)
Test#3 (Generic, ZIP)
122.82k
318.3k (x2.59)
Test#1 (Memory)
54.73k
27.2k (x0.5)
TOTAL
456.8k
935.66k (x2.05)

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
i9-13900K
Test#1 (Integers)
16.59k
31.93k (x1.92)
Test#2 (FP)
24.55k
31.54k (x1.28)
Test#3 (Generic, ZIP)
8.55k
18.8k (x2.2)
Test#1 (Memory)
26.28k
20.82k (x0.79)
TOTAL
75.97k
103.09k (x1.36)

Multithread

3900

i9-13900K
Test#1 (Integers)
181.33k
522.07k (x2.88)
Test#2 (FP)
261.01k
584.17k (x2.24)
Test#3 (Generic, ZIP)
126.62k
344.38k (x2.72)
Test#1 (Memory)
66.64k
25.36k (x0.38)
TOTAL
635.6k
1475.97k (x2.32)

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
i9-13900K
Test#1 (Integers)
16.16k
31.8k (x1.97)
Test#2 (FP)
23.75k
32.83k (x1.38)
Test#3 (Generic, ZIP)
9.32k
18.37k (x1.97)
Test#1 (Memory)
23.17k
21.13k (x0.91)
TOTAL
72.4k
104.13k (x1.44)

Multithread

3900

i9-13900K
Test#1 (Integers)
174.73k
532.97k (x3.05)
Test#2 (FP)
275.63k
596.78k (x2.17)
Test#3 (Generic, ZIP)
125.65k
342.96k (x2.73)
Test#1 (Memory)
44.17k
28.6k (x0.65)
TOTAL
620.18k
1501.31k (x2.42)

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
i9-13900K
Test#1 (Integers)
16.85k
58.6k (x3.48)
Test#2 (FP)
26.03k
33.29k (x1.28)
Test#3 (Generic, ZIP)
9.54k
17.77k (x1.86)
Test#1 (Memory)
22.55k
19.49k (x0.86)
TOTAL
74.97k
129.15k (x1.72)

Multithread

3900

i9-13900K
Test#1 (Integers)
229.41k
804.04k (x3.5)
Test#2 (FP)
292.81k
613.45k (x2.1)
Test#3 (Generic, ZIP)
128.48k
311.08k (x2.42)
Test#1 (Memory)
36.8k
34.5k (x0.94)
TOTAL
687.5k
1763.08k (x2.56)

Performance/W
3900
i9-13900K
Test#1 (Integers)
3529 points/W
6432 points/W
Test#2 (FP)
4505 points/W
4908 points/W
Test#3 (Generic, ZIP)
1977 points/W
2489 points/W
Test#1 (Memory)
566 points/W
276 points/W
TOTAL
10577 points/W
14105 points/W

Performance/GHz
3900
i9-13900K
Test#1 (Integers)
3920 points/GHz
10104 points/GHz
Test#2 (FP)
6054 points/GHz
5739 points/GHz
Test#3 (Generic, ZIP)
2218 points/GHz
3064 points/GHz
Test#1 (Memory)
5243 points/GHz
3360 points/GHz
TOTAL
17435 points/GHz
22268 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