From 9be3b226818b19b0bd4902119161eda10ff1dcf5 Mon Sep 17 00:00:00 2001 From: Rin Oliver Date: Thu, 18 Jun 2026 15:21:48 -0500 Subject: [PATCH 1/3] fix: correct CB R23 label to R24 in standard models table, clarify test conditions The standard models table was updated with macOS 15.1 / Xcode 16.1 / Cinebench R24 data but the column header still said CB R23. The full results table remains the original macOS 13.4.1 / CB R23 run. Pending: baseline section and full results table still need Matthew's updated numbers (tracked in DI ticket). Co-Authored-By: Claude Sonnet 4.6 --- iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx b/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx index dac8e890..7cee9f94 100644 --- a/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx +++ b/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx @@ -26,7 +26,7 @@ When evaluating Macs, the goal is to determine the optimal performance of a give **Benchmark Results** -All tests performed on macOS 13.4.1, unless otherwise specified. +The standard models table reflects tests performed on macOS 15.1 with Xcode 16.1 and Cinebench R24. The full results table reflects the original benchmark run on macOS 13.4.1 with Cinebench R23 and is included for reference. ## Benchmarking Baseline @@ -40,7 +40,7 @@ This machine is used as the baseline when referencing multipliers. In other word This table includes Standard models currently available as Bare Metal from MacStadium. -| Name | CPU | Xcode | GB6 CPU | GB6 GPU | CB R23 | RAM | SSD | Mac Model | +| Name | CPU | Xcode | GB6 CPU | GB6 GPU | CB R24 | RAM | SSD | Mac Model | |---|---|---|---|---|---|---|---|---| | S2.L | M2 Ultra
24C | 98.043s
(1.94x) | 21734
(2.12x) | 215160
(4.45x) | 1975.14
(3.20x) | 128GB | 2TB | Mac Studio | | S2.M | M2 Ultra
24C | 96.82s
(1.97x) | 21777
(2.13x) | 218001
(4.51x) | 1981.3
(3.30x) | 64GB | 2TB | Mac Studio | From 00e0c4efd66ccd8f6aa91b11a1d48d372dbfcf77 Mon Sep 17 00:00:00 2001 From: Rin Oliver Date: Thu, 18 Jun 2026 15:53:37 -0500 Subject: [PATCH 2/3] update: switch baseline to M2.M, drop Intel rows from full results Per Matthew Pulsipher: M1 was never re-tested against the newer benchmark suite, so M2.M is now the 1.00x baseline. Intel machines dropped from the full results table as performance is no longer a differentiating use case for that hardware. Co-Authored-By: Claude Sonnet 4.6 --- .../macstadium-bare-metal-mac-benchmarks.mdx | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx b/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx index 7cee9f94..f7675ca5 100644 --- a/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx +++ b/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx @@ -32,9 +32,9 @@ The standard models table reflects tests performed on macOS 15.1 with Xcode 16.1 This machine is used as the baseline when referencing multipliers. In other words, a machine with a multiplier of 1.4x is 40% more performant than the baseline machine for the specified test. Multipliers are shown below the raw score in each cell. -| Name | CPU | Xcode | GB6 CPU | GB6 GPU | CB R23 | RAM | SSD | Mac Model | +| Name | CPU | Xcode | GB6 CPU | GB6 GPU | CB R24 | RAM | SSD | Mac Model | |---|---|---|---|---|---|---|---|---| -| G5L | M1
8C | 123.885
(1.00x) | 8818
(1.00x) | 33843
(1.00x) | 7869
(1.00x) | 16GB | 2TB | Mac mini | +| M2.M | M2
8C | 190.33s
(1.00x) | 10244
(1.00x) | 48348
(1.00x) | 600.07
(1.00x) | 16GB | 1TB | Mac mini | ## Current standard models @@ -69,7 +69,3 @@ This table includes the full range of benchmarked hardware, including several ma | M2.M | M2
8C | 106.259s
(1.17x) | 10130
(1.15x) | 45902
(1.36x) | 8776
(1.12x) | 16GB | 1TB | Mac mini | | M2.S | M2
8C | 108.477s
(1.14x) | 10192
(1.16x) | 45891
(1.36x) | 8780
(1.12x) | 8GB | 256GB | Mac mini | | M1.M | M1
8C | 123.885s
(1.00x) | 8818
(1.00x) | 33843
(1.00x) | 7869
(1.00x) | 16GB | 1TB | Mac mini | -| G4B | i7 3.2
6C | 188.779s
(0.66x) | 6754
(0.77x) | 6631
(0.20x) | 7645
(0.97x) | 32GB | 512GB | Mac mini (2018) | -| G2 | Xeon
12C | 193.603s
(0.73x) | 5311
(0.60x) | 24788
(0.73x) | 8507
(1.08x) | 64GB | 1TB | Mac Pro (2013) | -| G2 | Xeon
6C | 264.804s
(0.47x) | 3862
(0.44x) | 24887
(0.74x) | 3862
(0.49x) | 32GB | 1TB | Mac Pro (2013) | -| G4A | i3 3.6
4C | 369.022s
(0.36x) | 3912
(0.44x) | 5806
(0.17x) | 3606
(0.46x) | 16GB | 512GB | Mac mini (2018) | From 364e6d7fce089b15e9be3729bec12c8a5ef6f89f Mon Sep 17 00:00:00 2001 From: Rin Oliver Date: Mon, 22 Jun 2026 15:04:40 -0500 Subject: [PATCH 3/3] fix: update Cinebench R23 reference to R24 in overview Co-Authored-By: Claude Sonnet 4.6 --- iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx b/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx index f7675ca5..5393a3c3 100644 --- a/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx +++ b/iaas/bare-metal-macs/macstadium-bare-metal-mac-benchmarks.mdx @@ -18,7 +18,7 @@ When evaluating Macs, the goal is to determine the optimal performance of a give * XcodeBenchmark * Geekbench 6 Multi-core * Geekbench 6 Compute - * Cinebench R23 + * Cinebench R24