The Leopard Experience at 867 MHz

I recently wrote about how nicely Leopard runs on PowerPC Macs. I just acquired an 867 MHz PowerBook G4. It needed some work (broken screen and no keyboard), but I have restored it and it works fine.

This is the slowest Mac to officially be supported by Mac OS X 10.5 "Leopard". For anyone wondering whether running a such a high spec modern OS on an older Mac works well, I say go for it. You will be surprised. I was.

I am no stranger to old Macs and pushing them to their limits, but this is by no means stretching it on an 867 MHz G4. Leopard really runs very smooth, very fast, and is more than just useable.

You could never run Microsoft Windows on a computer that meets the stated minimum system requirements and expect it to be useable. Windows 2000 requires a 133 MHz Pentium, but it is extremely slow even on a 200 MHz machine. Windows XP requires a 300 MHz PC, according to Microsoft, but that is just laughable.

The Mac Experience

Macs have been renowned for their longevity, and while they are initially expensive, they tend to outlast their PC counterparts, counterbalancing the cost. Minimum system requirements for a Mac OS do not mean working at a snail's pace. It seems Apple has thought its requirements through and thought of the user experience. While this might annoy users of Macs that don't meet the system requirements, it seems to ensure that even a bare minimum supported machine will run well.

I have been using my 867 MHz PowerBook G4 running Leopard for the past few days, and nothing lags on it. The Finder is snappy, CoverFlow is surprisingly sprightly. Running the usual bunch of apps - browsers, email clients, photo editing, and office suites - nothing seems to cause me any pain or stress the Mac out.

Heavy apps are usually where a low-end Mac will fall down, but Firefox 3 performs amazingly well, even with a whole host of tabs open, and even sites like YouTube work fine. Microsoft Office 2004 and OpenOffice open very quickly, and navigating through opening, editing, and saving documents is a breeze.

Even Photoshop CS and Ableton Live do not cause the machine to choke. I was expecting to see the "spinning wheel of death", but it has only popped up a few times for a few seconds.

Leopard Considerations

When considering Apple's minimum system requirements for Leopard, there is more than just the processor speed. You might think that performance shouldn't be that much different on an 867 MHz PowerBook to those from an 800 MHz PowerBook, but it is more than just processor power.

RAM has a lot of impact - the minimum requirement is 512 MB, and my 867 MHz PowerBook G4 has 768 MB - but there is even more to consider when it comes to Leopard. It is a very graphics heavy operating system, requiring a video card that supports Core Image and Quartz Extreme. Older Macs - even slightly older Macs than the minimum requirements - have older graphics capabilities, and this has a major effect on the stability and usability of Leopard.

I've run OS X 10.5 on a 400 MHz PowerBook G4 and a "Sawtooth" Power Mac G4, and while it worked okay - it certainly was useable for light tasks - there were a lot of graphical glitches, and CoverFlow was incredibly slow and occasionally hung the Finder.

Go for It

So to recap, if your Mac meets the minimum requirements for Leopard, go for it. If you have an 867 MHz PowerBook G4 or a 867 MHz "Quicksilver" Power Mac G4, let alone a Mac slightly above that, go for it. You will be surprised. It is quick.

There will be people saying, "No, stick with Tiger." I have run both on my PowerBook and on my eMac, and there isn't a lot of performance difference. I am not one of those people who claims that Leopard is faster than Tiger, because it isn't, but it certainly is on par with it.

Benchmark sites will say something different. If you run Xbench or Geekbench, they may give you a higher score running Tiger than Leopard on the same machine, but I don't think most users will see a huge difference.

See Low End Mac's Best Mac OS X 'Leopard' Prices for current deals.

Appendix: Tiger vs. Leopard Performance

Dan Knight, publisher, Low End Mac

Bear in mind that Geekbench, Xbench, and other benchmarks weight things differently. Geekbench looks almost exclusively at the CPU and memory path, while most other benchmarks include graphics and hard drive performance. Because of this, different benchmarks will show different results.

Another class of benchmarks isn't rooted in testing each part of the computer but instead in seeing how well it runs real world applications - these are the kind of things you'll find on Macworld and Bare Feats. Benchmark results are impacted by the CPU (kind, speed, number of cores, cache size), memory (amount, bandwidth), hard drive (buffer, rotation speed, throughput), graphics (GPU, video RAM, system bus, screen resolution), and the specific versions of the OS and benchmark software being used. More system memory, a faster hard drive, and a better video card can all improve overall system performance.

Two operating systems can only be objectively compared on exactly the same hardware, and benchmarks tend to show better scores under Leopard on Intel Macs, better scores on Tiger on PowerPC Macs. Despite the numbers, most users report that Leopard feels faster, and Leopard has been updated several times since the first Tiger vs. Leopard benchmarks were posted, so performance is probably better now than it was a year ago.

In the end, it's a matter of productivity: Will you be more productive with Leopard and its new features? Would moving to Leopard mean abandoning something that only works in earlier versions of the Mac OS (Classic Mode, for instance)?

• Leopard Performance (October 2007), Primate Labs, 2007.10.27. First reported Geekbench results compare Leopard and Tiger on two Macs.
• Leopard Faster than Tiger on Intel, Slower on PowerPC, and Possible Below 867 MHz, Dan Knight, Mac Musings, 2007.10.29. Looking at the first benchmarks comparing Mac OS X 10.4.x with 10.5.0.