Saturday, January 17, 2026

Budmark CPU Benchmark Results

Original benchmark by John Sokol (1998-2005)

Updated with modern hardware: January 2026

Overview

Budmark is a CPU benchmark based on brute-force searching for optimal error-correcting codes (ECC). The algorithm finds codewords with minimum Hamming distance 5, making it a pure integer/bit-manipulation workload that fits entirely in cache.

All results use unoptimized compilation (-O0) to measure raw CPU performance without compiler tricks.


Modern Hardware Results (2026)

Single-Core Performance

CPUArchitectureClockiter/secvs Xeon PII
Intel i5-4590x86-64 Haswell3.3GHz32,631251,000x
Pi 5 Cortex-A76ARMv82.4GHz18,454142,000x
Pi 4 Cortex-A72ARMv81.5GHz8,04962,000x
Pi Zero ARM1176ARMv61.0GHz1,32610,200x

Multi-Core Performance

CPUCoresSingleMultiScalingvs Xeon PII
Intel i5-4590432,631121,5003.73x935,000x
Pi 5 Cortex-A76418,45473,0003.96x562,000x
Pi 4 Cortex-A7248,04930,2003.75x232,000x
Pi Zero ARM117611,32610,200x

Historical Results (1998-2005)

From the original Budmark benchmark page.

CPUClockRun Time (s)iter/secEfficiencyOS
P43800MHz1.270.7971.6%WinXP Cygwin
P43000MHz1.570.6473.7%Linux 2.6.11
P42266MHz2.290.4466.7%FBSD 4.4
P41800MHz4.230.2445.5%Win2K Cygwin
P31150MHz3.060.3398.4%OBSD 3.1
VIA Eden1000MHz8.070.1242.9%Win2K Cygwin
P3866MHz3.990.25100.2%Slackware
Celeron766MHz4.670.2196.8%FBSD 4.6.2
AMD-K7550MHz7.280.1486.5%RedHat
Celeron533MHz6.720.1596.7%FBSD 4.6.2
Xeon PII450MHz7.700.13100%FBSD 3.0
AMD-K6450MHz8.470.1290.9%FBSD 2.2.7
Xeon PII400MHz8.690.1299.6%FBSD
AMD-K6350MHz10.310.1095.9%FBSD
Intel PII333MHz10.750.0996.8%FBSD 4.6.2
AMD-K6300MHz12.640.0891.3%FBSD
Cyrix GXm233MHz32.310.0346.0%FBSD
Pentium166MHz38.210.02654.6%FBSD 2.1.0
IBM Power2135MHz38.610.02666.4%AIX XLC -O2
Pentium133MHz47.900.02154.4%FBSD
Pentium120MHz52.920.01954.5%FBSD
486DX266MHz115.460.008745.4%FBSD 2.2.7
486DX66MHz153.380.006534.2%FBSD 3.1
48633MHz230.420.004345.5%FBSD 3.1
386DX40MHz537.780.001916.1%FBSD 3.1
38640MHz784.510.001311.0%FBSD 3.1
38616MHz1997.800.000510.8%FBSD 3.1

Efficiency Analysis

Efficiency measures work-per-clock-cycle, normalized to Xeon PII 450MHz = 100%.

CPUClockEfficiencyNotes
P3 866MHz866MHz100%Peak efficiency era
Xeon PII450MHz100%Baseline
P4 3800MHz3800MHz72%NetBurst penalty
P4 1800MHz1800MHz46%Early P4 very inefficient
i5-45903300MHz~34,000%Modern IPC gains
Pi 5 A762400MHz~26,000%ARM efficiency

Key observation: The Pentium 4 (NetBurst) architecture traded efficiency for clock speed. A P4 at 3.8GHz was only ~6x faster than a Xeon PII at 450MHz, despite having 8.4x the clock speed.

Modern CPUs have recovered efficiency through:

  • Deeper pipelines with better branch prediction
  • Larger caches (L1/L2/L3)
  • Out-of-order execution improvements
  • Better memory controllers

Raspberry Pi Comparison

ModelCPUClockCoresPriceMulti iter/secValue (iter/$/sec)
Pi ZeroARM11761.0GHz1$51,326265
Pi 4Cortex-A721.5GHz4$3530,200863
Pi 5Cortex-A762.4GHz4$6073,0001,217

Pi 5 offers best performance per dollar for compute workloads.

All Pi models show near-perfect multicore scaling (3.75-3.96x on 4 cores).


Test Commands

Single-core test

gcc -O0 -o ecc4_original ecc4.c -lm
time ./ecc4_original 100000

Multi-core test (4 cores)

time (./ecc4_original 100000 & ./ecc4_original 100000 & ./ecc4_original 100000 & ./ecc4_original 100000 & wait)

Equivalent workload to 1998 Xeon (7.7s)

# On i5-4590: ~245,000 iterations
time ./ecc4_original 245000

Summary

EraBest CPUiter/secImprovement
1988386 16MHz0.0005
1998Xeon PII 450MHz0.13260x
2005P4 3800MHz0.791,580x
2014i5-4590 (single)32,63165M x
2023Pi 5 (multi)73,000146M x
2014i5-4590 (multi)121,500243M x

A $60 Raspberry Pi 5 is 562,000x faster than a 1998 enterprise Xeon server.


Benchmark and original data: John Sokol, 1998-2026
https://www.dnull.com/cpubenchmark/budmark3.html

No comments: