Fleet chip bakeoff · 2026-07-22
M1 Airs stay useful for Remotion-class jobs.
Across the same Remotion compositions, M2 Max led on every short job, but HeavyMotion on M1 Air finished within about 25–30% of M2 Max wall-clock. That matters if you are planning idle MacBook Air supply for overnight render capacity.
Hosts: local-m2max (Apple M2 Max, 96 GB) · silver1 / silver2 (Apple M1, 16 GB).
Compositions: LightMotion 5 s · MediumMotion 10 s · HeavyMotion 15 s @ 1080p30.
Power: package samples succeeded on M2 Max; M1 Air power needs an interactive admin approval and was not captured over SSH.
Compositions: LightMotion 5 s · MediumMotion 10 s · HeavyMotion 15 s @ 1080p30.
Power: package samples succeeded on M2 Max; M1 Air power needs an interactive admin approval and was not captured over SSH.
| Composition | Frames | M2 Max | M1 Silver1 | M1 Silver2 | M2 speedup vs S1 |
|---|---|---|---|---|---|
| LightMotion | 150 | 3.47 s | 8.78 s | 8.66 s | 2.53× |
| MediumMotion | 300 | 5.98 s | 9.14 s | 8.54 s | 1.53× |
| HeavyMotion | 450 | 32.62 s | 42.93 s | 40.87 s | 1.32× |
Reading the gap
- Light jobs: cold Chromium + encode overhead dominate; more CPU/GPU headroom on M2 Max wins hard (~2.5×).
- Medium: gap shrinks to ~1.5× - still clearly faster on Max.
- Heavy (particles + blur): M1 Airs stay within ~25–30% of Max. Remotion is Chromium-bound.
Long kinetic check (3 minutes)
On LongKinetic3m (5400 frames), measured walls were 111.5 s (M2 Max), 133.7 s (Silver1), and 145.6 s (Silver2). Idle Air supply is usable for sellable-length Remotion work; do not assume only Max/Studio hosts matter.
Hosts: M1 Air viability → M1 host page · host waitlist.