Apple M4 Chip Benchmarked: 22% Faster Single-Core and 25% Faster Multi-Core Performance

[Minus Infinity]

Encoder was what he was looking for. M3 had a decoder as well.

Sorry too quick on the trigger. I thought it was new feature they added seeing it was highlighted.

 

[Guwapo77]

Sorry too quick on the trigger. I thought it was new feature they added seeing it was highlighted.

Bro, it got me too. I was reading a thread on Mac Rumors and I saw some folks going back and forth over that very topic, so I had to read into it.

 

[ikjadoon]

Every CPU uArch engineer uses Geekbench. There's nothing re: Apple or Arm "forcing" Geekbench to "recompile" for their software or "software ecosystem" nor "clout with developers". SME is an Arm-standard extension. We can analyze Apple's cores just like any x86 core. Apple isn't reinventing what a CPU does or how it works.

You'll never hear Intel and AMD make the excuse that "Geekbench is for Apple". In reality, Intel also uses Geekbench. AMD also uses Geekbench.

...


TL; DR: Anyone can complain, but CPU designers rely on industry-standard benchmarks like Geekbench for a reason: reproducibility, relevance to consumers, and excellent for cross-OS uArch tests.

There is nothing niche about Geekbench and there's no specific bias with Arm CPUs when using Geekbench.

FYI, Geekbench 6 is a CPU core-only benchmark. It uses zero accelerators, e.g., nothing except the CPU instructions themselves.

https://www.geekbench.com/doc/geekbench6-benchmark-internals.pdf

Adding more evidence to the "every CPU designer on Earth takes Geekbench as a ground source of truth on CPU uArch design and 1T perf" take:

AMD's slides particularly show how well GB6 correlates with mainstream consumer 1T perf. After SPEC, Geekbench is genuinely a good benchmark (representative of its intended use cases; relatively precise; works cross-platform; dead simple to use).

//

Not to be trollish, but it remains surprising to me why so many commenters--across every forum--innately distrust Geekbench, but virtually every client CPU uArch company relies on Geekbench prodigiously and many heap praise on it for its representative performance modelling, reliability and precision across multiple platforms, and ease of us.

Intel Lunar Lake: perf estimated via GB6
AMD Zen5: perf estimated via GB6
Qualcomm NUVIA: perf estimated via GB6
Arm Cortex-X925: perf estimated via GB6
Some commenters: "GB6 is so trash, it doesn't correlate to anything. Worthless benchmark that favors [vendor X], I'm sure."

Surely, if GB was as biased or unrepresentative or not predictive as many claim it is, it would've left the arena a long time ago.

 

[star-affinity]

Apple will never do front facing USB slots on this device. the whole point of Apple is image/clean look. I am alright with this.

for $499 a m4 Mac Mini is going to be great bang for buck regarding work stuff.

The Mac Studio has front facing USB-C ports and an SD card slot, so I don't see why front facing ports should never happen on the Mac mini.