PowerBook vs. Mac mini -- a Zope 3 benchmark ============================================ A couple of weeks ago my mom's and dad's computer broke down. It's an iMac G3 400. You know, one of those old coloury things. In fact, this one's green (I think Apple called it *lime*.) Probably five years ago we bought this machine used off of eBay. At the time, it was perhaps two years old, perhaps a little more. Given excellent Linux support on the G3 Mac family (I used to happily run Linux on my iBook G3), it was a *very* cheap all-in-one machine that's just perfect for my parents desk. Add a user-friendly KDE desktop, a Mozilla suite and OpenOffice and you got a system that even my mom could use -- and she did, for the past five years. Now the CRT screen broke and it's time to move on. The timing couldn't be better. I just returned from China and Apple just released a new Mac mini with Intel processors. So, I went to the Apple store the other day to get my folks one of them new Mac minis. My parents only need to perform standard office tasks: Web, email, text and spreadsheet processing. So, I figured the single core version would be enough. It's also more than 200 euros cheaper than the dual core one. Sure it doesn't have the DVD burner and the big harddrive, but who needs that when you're just using the box as an advanced typewriter anyways? Naturally, I had to test the machine before I passed it on to my folks. First impression ---------------- I have to say, the Mac mini is great. It's so small and quiet that it fits perfectly on any desk. Heck, you need more room just to operate a mouse than this thing takes away. That alone makes it impressive and combined with the iRemote and FrontRow, it's a pretty good show-off for your friends. It *is* quite expensive. You can get more hardware for 600 euros. But as always with Apple, it's not the hardware alone that makes a good product. Frankly, I'm tired of administrating Linux boxes for my family. Sure, it's much better than giving them Windows and see them trash an installation every so many months that you have to come and reinstall the whole system. But, even given the vast UI improvements over the last years, it's still a pain to use when you're Joe Average User. With Mac OS X, however, I'm even confident enough that I'll make my dad an Administrator of the system. And I won't see any problem for my parents to hook up their new digital camera now or for my sister to make use of her newest toy, an iPod nano. Apple makes it so simple, and that is worth some money to me. Back to the mini. It doesn't feel incredibly fast when working in the OSX environment. In fact, I was a bit disappointed at first because I expected to be blown away by its speed. But really what did I expect? It was "just" a single processor at 1.5 GHz. For that, it works very well. It is still easily workable when all five family members are logged on and are running Safari, Mail, OpenOffice 2 etc each. The benchmark ------------- Of course, the Mac mini felt faster than my PowerBook which isn't even two years old. I just wondered *how much* faster it really was. The PowerBook is quite sufficient for every day jobs, it's Zope 3's 6800 unit tests and 524 functional tests that are executed painfully slow to run. So, obviously I had to know how fast the Mac mini would run the Zope 3 test suite. And here are the results, along with some general hardware comparison:: ============== =================== ======================= Powerbook Mac mini ============== =================== ======================= processor PowerPC G4 1.33 GHz Intel Core Solo 1.5 GHz RAM 1 GiB 512 MiB harddisk 80 GB ATA 5400 rpm 60 GB SATA 5400 rpm -------------- -------------------------------------------- OS Mac OS X 10.4.5 with latest security patches python 2.4.2 from darwinports zope 3 trunk, r66263 -------------- -------------------------------------------- pystone 1.1 19800 py/s 34700 py/s -------------- ------------------- ----------------------- compile[1] 77 s 52 s startup[2] 63 s 25 s 2nd startup[3] 21 s 8 s -------------- ------------------- ----------------------- unit tests[4] 344 s 173 s ftests setup 25 s 11 s run 610 s 245 s ============== =================== ======================= .. [1] ``make`` command, measured via Unix ``time``, real time shown. .. [2] ``z3.py`` command after having done "something else" (Zope 3 files not in disk cache); time measurement by Zope 3 startup machinery, real time shown. .. [3] ``z3.py`` command right after having terminated the first one; time measurements as above. .. [4] ``test.py -p --all`` command, time measurement by the test runner. Wow, that message is clear. The Mac mini is basically twice as fast as the PowerBook! Except when you start up Zope 3 for the first time since that is a disk-heavy operation and just as slow on both systems. The rest of the numbers aren't to be taken as scientific results. I rounded all times to the nearest second and the pystones to 100 pystones/s because anything else would be insignificant. Now I just wonder: how fast will it all be on the MacBook? Well, I won't find out any time soon because I'm not getting one... but that Mac mini doesn't sound so bad. It doesn't sound so bad at all.