Benchmarking the Mac Mini 2009 2.26 GHz (Part 1)

icerabbit on March 22nd, 2009

[Update: Part 2 with some external drives is posted here ]

Quite a number of people wonder if it is worth upgrading the Mac mini 2009 from its default Core 2 Duo running at 2 GHz to 2.26 GHz (13% CPU increase), so below are my first benchmark results.

I hadn’t see any 2.26 GHz benchmarks yet, when I decided to order my mini with this upgrade. I figured it was worth it since I was going to save money on doing the other hardware upgrades myself. Upgrading the CPU, memory & hard drive kicks a mini over $1000. 

Note that I consider 2GB pretty much a mandatory upgrade, so it enables both memory channels and unlocks the shared graphic memory from 128MB to 256MB. The jury is still out, I think, about a 7200 rpm drive.  Generally the hard drive is the slowest component in a computer (ignoring the optical drive) and a 7200 rpm drive should be faster than a 5400 rpm drive. But, other technical details come into play like amount of cache, # of platters, platter density, … and so you can see a great 5400 rpm drive hold its own against an average 7200 rpm model. Research is key.  The PowerBook greatly benefits from the 100GB 7200 rpm drive over the default one, but maybe Apple made a good hard drive choice here. 

Note: Geekbench 32 bit demo mode. 

Mac mini 2009 2.26 GHz – 1 GB RAM - OS X 10.5.6 – 1st Boot * 

Geekbench 32: 3007 — Integer: 2428 — Floating point: 4343 — Memory: 2410 — Stream: 1558

Xbench: 124.57 — CPU: 142.52 — Thread Test: 307.03 — Memory Test: 154.54 — Quartz Graphics Test: 170.22 — OpenGL Graphics Test: 110.46 — User Interface: 248.38 — Disk test: 48.79 

* I noticed after completion that Apple had a software update process running in the background that was pulling several hundred KB/s of updates, so that may have compromised the result.

Mac mini 2009 2.26 GHz – 1 GB RAM - OS X 10.5.6 – Restarted

Geekbench 32: 3032 — Integer: 2448 — Floating point: 4413 — Memory: 2413 — Stream: 1486

Xbench: 178.53  – CPU: 136.81 — Thread Test: 321.07 — Memory Test: 156.73 — Quartz Graphics Test: 179.30 — OpenGL Graphics Test: 137.04 — User Interface: 244.39 — Disk test: 48.52 

Upgrading the mini is not for the faint of heart, but I did it. More on that in my upgrade story. 

Mac mini 2009 2.26 GHz – 4 GB RAM - OS X 10.5.6 – Restarted

Geekbench 32: 3081 – Integer: 2492 — Floating point: 4378 — Memory: 2476 — Stream: 1821

Xbench: 130.57  – CPU: 137.74 — Thread Test: 252.28 — Memory Test: 174.42 — Quartz Graphics Test: 185.21 — OpenGL Graphics Test: 136.94 — User Interface: 285.55 — Disk test: 48.90

I had no idea why the second Xbench score was significantly higher the second time around with 1GB, only to see it drop back to a still improved level with 4GB. A fourth & fifth result were 131. Actually, the only difference that I am aware of was that the initial monitor topped out at 1024 x 768 pixels, versus the mini’s monitor now being 1920 x 1080. Back to the original 15″ monitor: Xbench 132. 

Also noteworthy:

Default installation takes up close to 17GB with the OS & iLife ‘09. 

Boot time: 4 seconds to the “bong” and 35 seconds to desktop fully loaded with custom menu bar (iStat Menu), Wifi on DHCP, BlueTooth on. 

Shut down time: I have seen it shut down as fast as 5 seconds. Another time 32 seconds, where the display goes blank in 3-4 seconds, but it takes another half a minute for the mini’s power light to extinguish. 

All in all I am very pleased with the mini’s performance.

6 Responses to “Benchmarking the Mac Mini 2009 2.26 GHz (Part 1)”

  1. I too, had weird results on my mac mini 1.83GHz:

    From 5/08 my XBench 1.3 test with 1GB ram was 100.64
    From 3/09 my XBench 1.3 test with upgraded 4GB ram was 46.64

    Do not see any degradation with 4GB ram – just notice that I don’t run out of ram when doing a lot of surfing on the internet.

    Best, Brad

  2. Hi Brad.
    Does XBench still give a lower result on additional tests?
    That second number is very low, I think, for your system.
    Not something that would even be caused by slower memory.

  3. Interesting comment re 5400 v 7200 rpm drives; my next Mac is likely to be a portable (my first since the venerable PowerBook 100) and I was wondering whether it would be worth the upgrade to a faster drive. I’d assumed the performance difference would be notable, at the cost of worsening battery life (less of a concern with a desktop replacement). Another thing I’m curious about is the FireWire 800 port. I have a LaCie enclosure with FW 800 ports (only 150 GB drive, shame, @ 7200 rpm) and wondered if booting from that might actually make for a faster machine than using an internal 5400 rpm drive. Any ideas on that?

  4. fw 800 would never be as fast as the internal esata buss, google the topic for external HD booting where you can run and exteral esata cable.

  5. Thanks for the comment.

    I agree that the internal bus should be the fastest option, but regardless of bus type, it is still the actual hard drive itself that is the slowest factor. Hard drives only saturate old bus types, and so I am interested in testing if there is a benefit … or if it no longer holds true.

    I recall seeing an article about a mini with sata connected to an external drive. I am unfortunately not willing to dremel a section out of my system. At least not while it is under warranty.

Trackbacks/Pingbacks

  1. Benchmarking the Mac mini 2.26 (Part 2) | A Brugian in the US

Leave a comment - Voeg een reactie toe