What's that old saying about the best-laid plans of Mice and Macs? Well, my original plan was to post one more diary is this series, a catch-all that included a round-up of third party add-ons I'd tested, my final impressions of the product, and the benchmarks of the 400-MHz version of the PowerBook G4.
But, that's before I reviewed those benchmarks. They were so interesting - and surprising -- I wanted to get them up right away. After all, the money you save once you see these numbers might be your own!
From the standpoint of physical design, there are no differences between the 400 and 500-MHz TiBook. It offers the same thin, sleek exterior, beautiful wide screen, ports and snappy keyboard. Under the skin, the TiBook 400 has 128 MB of RAM versus 256 MB on the TiBook 500, a 10 GB drive versus a 20 GB drive, and, of course, a 100-MHz less in CPU clockrate.
I expected to see roughly 10 percent to 20 percent difference in performance, depending if you were looking at CPU-intensive tasks or SpeedMark scores.
I was surprised on both counts.
Sorry, this diary only features one piece of art, but what a very informative piece of art it is:
As you can see, in tests of Photoshop filters that are Altivec (Velocity Engine) -savvy, there is no distinguishable difference in speed between a 400 and 500-MHz G4 system. That means the delta in performance on Altivec-related benchmarks (except for Lighting Effects), is less than one second! This certainly might give buyers who are looking to get the fastest Photoshop speeds reason to consider the cheaper PowerBook configuration. Why pay more for a less-than-a-second difference?
Those users looking for the best overall system performance would be well advised to consider the SpeedMark scores, which give the 500-MHz machine the edge in overall performance. No, the big surprise was that in SpeedMark, there was no difference in performance between a G4 400 and a PowerBook G3 500, the recently discontinued high-end of the PowerBook line. Remember, SpeedMark uses a suite of applications running common tasks to come up with a single measure of speed. The tasks run the gamut from CPU-intensive to disk-intensive, and are designed to reflect an overall speed doing everyday things.
And, in the Photoshop RGB to CMYK test, a CPU-intensive task that is not Altivec-savvy, the PowerBook G3 500 is significantly faster than the PowerBook G4 400!
After the Lab analysts helped me off the floor (I'd fainted when I read this last set of test results), we discussed why this was happening. And, after some consideration, we realized this result did in fact make sense. Take away Altivec, and a G4 is basically a G3 in terms of performance. So, it makes sense that a 500-MHz G3 would be faster than a 400-MHz G4 running CPU tasks that are not Altivec-savvy.
On the Quake III test, the difference in speed is, again, very modest, as Quake primarily tests graphics performance and both units use the same ATI RAGE Mobility chipset. The SoundJam test gave the nod to the TiBook 500, where Altivec-enhanced performance really pays off. By the way, for those of you who were wondering why the TiBook 500 beat the Power Mac G4 533 in the SoundJam test, it's because the TiBook's new DVD-ROM mechanism offers faster throughput than the mechanism in the 533 desktop machine.
So, now for the big question, which is the best value? Well, here's what it comes down to: Pricing out the cost of upgrades at the Apple online store, the price for adding an additional 128 Mb of RAM to the PowerBook G4 400 is $300. Upgrading to a 20 MB hard drive adds another $200. That means it will cost you $500 to make a TiBook 400 feature-equivalent with a TiBook 500. And, keep in mind, you can get much better prices on RAM from a third-party supplier: Outpost.com offers the same RAM upgrade for $70.95.
So, let's say you buy your new PowerBook G4 400 from Apple with a 20 GB hard drive, and opt for the cheaper RAM from Outpost -- the difference in price is only $271. (Outpost doesn't charge for overnight shipping or tax, so there's no additional cost buying your RAM from them.) That means if you purchase a TiBook 500, you're paying $629 extra for what generally amounts to a modest difference in performance.
That TiBook 400 is beginning to sound like a pretty sweet deal!
One last thing: Tonight while I was writing up these results, I installed an Airport card in the TiBook 400 to see if it gave me trouble when getting at the slot under the bottom cover. I'm happy to report things went a bit better than they did with the 500-Mhz machine (see the Diary, Day Four). The screws were still a bit tight in the chassis, but with a little caution the cover went back on with only a minimum of hassle. And only one screw tried to go in at an angle (I was ultimately able to get it to go in straight.) I'm now convinced that if someone at Apple hadn't installed the card on the TiBook 500 and, apparently in a rush, tightened a couple of screws in at an angle, I would have had far fewer problems with that unit, as well.
I did encounter one new problem with the TiBook 400: I had a heck of time getting the battery out of the machine. There was something wrong with the latch that's supposed to pop up the battery - it didn't work. I ultimately had to use a flat-head screwdriver like a lever to get the battery out.
Oh, well, nothing's ever perfect, is it?
Tune in next week for the last (really!) installment of the PowerBook G4 test diaries, where I share my final thoughts on Apple's new portables and bestow the mouse rating.