Benchmarking WBPP on Apple Silicon – M1 iMac vs M4 Pro Mac Mini
Processing the Anteater Nebula (LDN 1657A Region)
As part of my ongoing project on the Anteater Nebula, I decided to benchmark the performance of PixInsight’s Weighted Batch Preprocessing (WBPP) using my full dataset—over 50 hours of narrowband and broadband exposures—on two different Apple Silicon machines.
The comparison?
My older M1 iMac with 16GB RAM
My new M4 Pro Mac Mini with 64GB RAM
Both machines are running macOS Tahoe 26.1 with identical PixInsight versions and WBPP configs.
WBPP Dataset
Target: LDN 1657A – Anteater Nebula
Total Integration: 50h 31m (Ha, OIII, SII, RGB, Luminance)
Image Scale: 9576 × 6388
Drizzle: 2x, 0.60 pixel fraction, square
WBPP Version: 2.8.9
Fast Integration: Disabled
All subs: bin 1x1, mono
📷 image.png
Performance Comparison
📷 image.png
That’s a 4x improvement in total processing time, using the exact same files, calibration structure, and PixInsight configuration.
Observations
The M4 Pro breezed through the stack with no signs of memory pressure, even during the drizzle integration steps.
The M1 iMac, while still capable, slowed dramatically during calibration and integration, often maxing out its RAM and relying on swap memory.
For workflows involving multi-filter datasets or high-res CMOS frames, the upgrade is night and day.
This wasn’t just about speed…it was about unlocking headroom for future projects. If you’re processing large datasets on Apple Silicon and thinking about scaling up your imaging workflow, the M4 Pro + 64GB config is an exceptional leap forward. Really glad I purchased it.
