Geek Patrol


Aperture System Requirements

With the release of Aperture 1.5, Apple has added, among other things, support for more graphics cards, including the Intel GMA 950 and the nVidia GeForce FX 5200 Ultra. This means Aperture will run on any Intel-based Mac, as well as any Power Mac G5. Woo hoo!

What really caught my eye, though, was the minimum memory requirements for Aperture:

  • 1GB of RAM
  • 2GB of RAM required for the Mac Pro

Um, what? Why would the Mac Pro require 2GB of RAM? Sometimes I just can’t figure out what goes on inside of Apple.


32-bit vs 64-bit Performance Under Mac OS X

One of the assertions I keep hearing is that with the transition from 32-bit to 64-bit computing applications will receive an increase in performance “for free” since processors will be working with 64 bits, instead of 32 bits, at a time (provided, of course, applications are re-compiled for 64-bit platforms).

Now that Geekbench is available for both 32- and 64-bit processors on Mac OS X, I thought I’d see if that assertion is correct. I’ve compared performance on the three 64-bit processors currently available for Mac OS X; the PowerPC G5, the Intel Xeon, and the Intel Core 2 Duo.

Setup

Here are the configurations of the test machines:

  • Mac Pro

    • Intel Xeon 5150 @ 2.66GHz
    • 2048MB RAM
    • Mac OS X 10.4.7 (Build 8K1124)
    • Geekbench 2006 (Build 208)
  • iMac (Late 2006)

    • Intel Core 2 Duo @ 2.0GHz
    • 1024MB RAM
    • Mac OS X 10.4.7 (Build 8K1106)
    • Geekbench 2006 (Build 208)
  • Power Mac G5

    • PowerPC G5 @ 2.0GHz (two processors)
    • 1024MB RAM
    • Mac OS X 10.4.7 (Build 8J135)
    • Geekbench 2006 (Build 208)

I’ve only included the scores for single-threaded tests (since I think they’re the most relevant when comparing 32- and 64-bit performance on the same machine). I’m using the baseline score (where a score of 100 is equivalent to the performance of a Power Mac G5 at 1.6GHz), where higher is better. I’ve also computed the 64-bit score for each machine as a percentage of the machine’s 32-bit score.

Mac Pro Performance

Overall Score

Mac Pro
(32-bit)
Mac Pro
(64-bit)
Overall Score 344.8 365.2
(105.9%)

Integer Performance

Benchmark Mac Pro
(32-bit)
Mac Pro
(64-bit)
Emulate 6502
single-threaded scalar
162.8218.0
(133.9%)
Blowfish
single-threaded scalar
232.8205.2
(88.1%)
bzip2 Compress
single-threaded scalar
223.5277.4
(124.1%)
bzip2 Decompress
single-threaded scalar
251.9300.4
(119.3%)

Floating Point Performance

Benchmark Mac Pro
(32-bit)
Mac Pro
(64-bit)
Mandelbrot
single-threaded scalar
179.9180.0
(100.1%)
Dot Product
single-threaded scalar
362.0364.1
(100.6%)
Dot Product
single-threaded vector
153.5127.2
(82.9%)
JPEG Compress
single-threaded scalar
161.0195.7
(121.6%)
JPEG Decompress
single-threaded scalar
154.9199.9
(129.1%)

Memory Performance

Benchmark Mac Pro
(32-bit)
Mac Pro
(64-bit)
Read Sequential
single-threaded scalar
354.3356.7
(100.7%)
Write Sequential
single-threaded scalar
631.3423.1
(67.0%)
Stdlib Allocate
single-threaded scalar
279.0357.1
(128.0%)
Stdlib Write
single-threaded scalar
124.3116.1
(93.4%)
Stdlib Copy
single-threaded scalar
234.9252.1
(107.3%)

Stream Performance

Benchmark Mac Pro
(32-bit)
Mac Pro
(64-bit)
Stream Copy
single-threaded scalar
199.4204.2
(102.4%)
Stream Copy
single-threaded vector
197.1198.2
(100.6%)
Stream Scale
single-threaded scalar
217.6208.7
(95.9%)
Stream Scale
single-threaded vector
196.4195.6
(99.6%)
Stream Add
single-threaded scalar
188.4210.7
(111.8%)
Stream Add
single-threaded vector
184.7218.2
(118.1%)
Stream Triad
single-threaded scalar
147.0210.0
(142.9%)
Stream Triad
single-threaded vector
183.7173.9
(94.7%)

Mac Pro Summary

Overall performance in 64-bit mode is 5% higher than overall performance in 32-bit mode. However, a number of benchmarks that were slower in 64-bit mode than in 32-bit mode (like the Blowfish and Write Sequential benchmarks).

iMac Performance

Overall Score

iMac
(32-bit)
iMac
(64-bit)
Overall Score 205.2 221.5
(107.9%)

Integer Performance

Benchmark iMac
(32-bit)
iMac
(64-bit)
Emulate 6502
single-threaded scalar
122.3164.0
(134.1%)
Blowfish
single-threaded scalar
175.0153.5
(87.7%)
bzip2 Compress
single-threaded scalar
168.5209.7
(124.5%)
bzip2 Decompress
single-threaded scalar
212.9227.7
(107.0%)

Floating Point Performance

Benchmark iMac
(32-bit)
iMac
(64-bit)
Mandelbrot
single-threaded scalar
135.1135.1
(100.0%)
Dot Product
single-threaded scalar
271.5273.2
(100.6%)
Dot Product
single-threaded vector
113.2115.0
(101.6%)
JPEG Compress
single-threaded scalar
120.7147.2
(122.0%)
JPEG Decompress
single-threaded scalar
116.1154.8
(133.3%)

Memory Performance

Benchmark iMac
(32-bit)
iMac
(64-bit)
Read Sequential
single-threaded scalar
308.4307.5
(99.7%)
Write Sequential
single-threaded scalar
416.7439.5
(105.5%)
Stdlib Allocate
single-threaded scalar
208.6273.4
(131.1%)
Stdlib Write
single-threaded scalar
104.3104.7
(100.4%)
Stdlib Copy
single-threaded scalar
218.9221.1
(101.0%)

Stream Performance

Benchmark iMac
(32-bit)
iMac
(64-bit)
Stream Copy
single-threaded scalar
170.7178.1
(104.3%)
Stream Copy
single-threaded vector
161.0159.1
(98.8%)
Stream Scale
single-threaded scalar
183.2176.0
(96.1%)
Stream Scale
single-threaded vector
160.8160.4
(99.8%)
Stream Add
single-threaded scalar
159.0192.5
(121.1%)
Stream Add
single-threaded vector
176.2179.0
(101.6%)
Stream Triad
single-threaded scalar
159.7187.8
(117.6%)
Stream Triad
single-threaded vector
141.7148.9
(105.1%)

iMac Summary

Despite the fact that the Core 2 Duo and the Xeon share the same underlying architecture, the Core 2 Duo’s 64-bit performance is better than the Xeon’s 64-bit performance; overall performance for the Core 2 Duo is up 7% (compared to 5% for the Xeon). Plus, the only benchmark that was significantly slower in 64-bit mode was the Blowfish benchmark.

Power Mac Performace

Overall Score

Power Mac G5
(32-bit)
Power Mac G5
(64-bit)
Overall Score 154.9 140.1
(90.4%)

Integer Performance

Benchmark Power Mac G5
(32-bit)
Power Mac G5
(64-bit)
Emulate 6502
single-threaded scalar
125.1100.0
(79.9%)
Blowfish
single-threaded scalar
124.789.0
(71.4%)
bzip2 Compress
single-threaded scalar
156.5110.8
(70.8%)
bzip2 Decompress
single-threaded scalar
108.7106.0
(97.5%)

Floating Point Performance

Benchmark Power Mac G5
(32-bit)
Power Mac G5
(64-bit)
Mandelbrot
single-threaded scalar
125.2129.8
(103.7%)
Dot Product
single-threaded scalar
112.3112.8
(100.4%)
Dot Product
single-threaded vector
125.542.0
(33.5%)
JPEG Compress
single-threaded scalar
122.0105.9
(86.8%)
JPEG Decompress
single-threaded scalar
129.6107.5
(82.9%)

Memory Performance

Benchmark Power Mac G5
(32-bit)
Power Mac G5
(64-bit)
Read Sequential
single-threaded scalar
133.9130.1
(97.2%)
Write Sequential
single-threaded scalar
145.9161.2
(110.5%)
Stdlib Allocate
single-threaded scalar
101.993.2
(91.5%)
Stdlib Write
single-threaded scalar
129.7131.1
(101.1%)
Stdlib Copy
single-threaded scalar
134.2124.7
(92.9%)

Stream Performance

Benchmark Power Mac G5
(32-bit)
Power Mac G5
(64-bit)
Stream Copy
single-threaded scalar
132.9127.9
(96.2%)
Stream Copy
single-threaded vector
129.2122.8
(95.0%)
Stream Scale
single-threaded scalar
129.9127.5
(98.2%)
Stream Scale
single-threaded vector
129.6131.1
(101.2%)
Stream Add
single-threaded scalar
127.4129.3
(101.5%)
Stream Add
single-threaded vector
130.8140.7
(107.6%)
Stream Triad
single-threaded scalar
134.5129.7
(96.4%)
Stream Triad
single-threaded vector
137.1139.5
(101.8%)

Power Mac Summary

Overall performance is down 10% in 64-bit mode. Hardly any tests are appreciably faster in 64-bit mode, and several are noticeably slower (such as most of the integer tests, as well as the dot product test).

Conclusion

It turns out the assertion that software runs faster in 64-bit mode than 32-bit mode is both correct and incorrect; Geekbench runs faster in 64-bit mode on Intel-based Macs, but slower on PowerPC-based Macs. I find this incredibly surprising.

On Intel-based Macs, most of the benchmarks that are slower in 64-bit mode are benchmarks that perform bit operations on 32-bit integers, where the compiler has to emit extra instructions to preserve the semantics of 32-bit arithmetic while using 64-bit registers.

However, extra instructions don’t explain the surprising performance hit PowerPC-based Macs experience in 64-bit mode. I haven’t had a chance to investigate it, but compiler quality could be a factor; the 64-bit PowerPC is a somewhat exotic platform, and GCC might not be generating great code for it.

I don’t think the performance hit in 64-bit mode on PowerPC-based Macs is really something to be concerned about; I think that when 64-bit applications become mainstream, most users will have switched to Intel-based Macs (where 64-bit performance isn’t a concern).

Update

There’s an interest comment over on MacSlash suggesting why 64-bit performace (compared to 32-bit performace) is better on x86 than PPC:

As someone who used to work at AMD which designed the x86-64 architecture: – 16 integer pipe registers versus 8 in 32 bit mode (of which 6 get used) – Carefully designed CISC so that 64-bit mode takes only 10% more space than 32 bit mode. This is important because the main bottleneck in modern systems is memory speed (hence the constant increase in cache sizes) PowerPC: – no increase in registers – much larger code size increase, although I can’t find exact figures.


Viewsonic Announces iPod Accessory

Viewsonic has announced a new 23″ LCD widescreen display which features a USB2 hub, integrated speakers, an 8-in-1 card reader and a freaking iPod dock.

Here’s a pretty picture:

Viewsonic iPod LCD

I’ve no idea what to say. I’m simultaneously stunned and completely unsurprised.


Geekbench 2006 Updated (Build 208)

The latest update to Geekbench 2006 (build 208 for those of you keeping score at home) is now available for download. With this release, Geekbench 2006 is available as a 32- and 64-bit benchmark on Mac OS X for both Intel- and PowerPC-based Macs. This means that Geekbench is the first ready-to-run 64-bit benchmark for Mac OS X. Woo hoo!

Check out the release notes for all the details.


Mac mini (1.66GHz and 1.83GHz) Benchmarks

Apple released updated Mac minis a couple of weeks ago. They weren’t incredibly exciting updates (just a minor processor upgrade with no new models or functionality), but we thought we’d take a look at the performance of the old high-end Mac mini (with an Intel Core Duo @ 1.66GHz) against the new high-end Mac mini (with an Intel Core Duo @ 1.83GHz).

Update: If you’re curious how PowerPC-based Mac mini performance compares to Intel-based Mac mini performance, take a look at these benchmarks I published back in March.

Setup

Here is the configuration of our two test machines:

  • Mac mini (Early 2006)
    • Intel Core Duo @ 1.66GHz
    • 1024 MB RAM
    • Mac OS X 10.4.6 (Build 8I1119)
    • Geekbench 2006 (Build 200)
  • Mac mini (Early 2006)
    • Intel Core Duo @ 1.83GHz
    • 2048 MB RAM
    • Mac OS X 10.4.7 (Build 8J2135)
    • Geekbench 2006 (Build 200)

As always, we’re using the baseline score (not the raw scores) from each benchmark. We’ve also computed the Mac mini Core Duo 1.83GHz score as a percentage of the Mac mini Core Duo 1.66GHz score (to make the comparison easier). For all scores, higher is better.

Overall Score

Core Duo 1.66GHz Core Duo 1.83GHz
Overall Score 155.5 164.4
(105.7%)

Integer Performance

Benchmark Core Duo 1.66GHz Core Duo 1.83GHz
Emulate 6502
single-threaded scalar
76.484.1
(110.1%)
Emulate 6502
multi-threaded scalar
152.6167.9
(110.0%)
Blowfish
single-threaded scalar
135.4149.2
(110.2%)
Blowfish
multi-threaded scalar
270.1295.9
(109.6%)
bzip2 Compress
single-threaded scalar
123.2135.7
(110.1%)
bzip2 Compress
multi-threaded scalar
244.9268.4
(109.6%)
bzip2 Decompress
single-threaded scalar
113.0124.3
(110.0%)
bzip2 Decompress
multi-threaded scalar
231.4252.3
(109.0%)

Floating Point Performance

Benchmark Core Duo 1.66GHz Core Duo 1.83GHz
Mandelbrot
single-threaded scalar
117.9129.7
(110.0%)
Mandelbrot
multi-threaded scalar
235.3258.7
(109.9%)
Dot Product
single-threaded scalar
255.3280.1
(109.7%)
Dot Product
multi-threaded scalar
367.6413.4
(112.5%)
Dot Product
single-threaded vector
87.197.0
(111.4%)
Dot Product
multi-threaded vector
167.0183.8
(110.1%)
JPEG Compress
single-threaded scalar
91.7101.0
(110.1%)
JPEG Compress
multi-threaded scalar
183.5201.6
(109.9%)
JPEG Decompress
single-threaded scalar
84.693.5
(110.5%)
JPEG Decompress
multi-threaded scalar
165.9181.8
(109.6%)

Memory Performance

Benchmark Core Duo 1.66GHz Core Duo 1.83GHz
Read Sequential
single-threaded scalar
192.6199.0
(103.3%)
Read Sequential
multi-threaded scalar
171.2171.1
(99.9%)
Write Sequential
single-threaded scalar
121.9109.3
(89.7%)
Write Sequential
multi-threaded scalar
176.6177.0
(100.2%)
Stdlib Allocate
single-threaded scalar
161.8177.7
(109.8%)
Stdlib Allocate
multi-threaded scalar
42.055.6
(132.4%)
Stdlib Write
single-threaded scalar
103.5103.9
(100.4%)
Stdlib Write
multi-threaded scalar
115.5115.7
(100.2%)
Stdlib Copy
single-threaded scalar
204.4215.7
(105.5%)
Stdlib Copy
multi-threaded scalar
194.1210.0
(108.2%)

Stream Performance

Benchmark Core Duo 1.66GHz Core Duo 1.83GHz
Stream Copy
single-threaded scalar
121.6131.1
(107.8%)
Stream Copy
multi-threaded scalar
172.4177.4
(102.9%)
Stream Copy
single-threaded vector
114.6116.1
(101.3%)
Stream Copy
multi-threaded vector
158.4158.0
(99.7%)
Stream Scale
single-threaded scalar
131.2127.9
(97.5%)
Stream Scale
multi-threaded scalar
182.5182.8
(100.2%)
Stream Scale
single-threaded vector
112.4112.5
(100.1%)
Stream Scale
multi-threaded vector
156.4154.7
(98.9%)
Stream Add
single-threaded scalar
100.8104.0
(103.2%)
Stream Add
multi-threaded scalar
159.3164.6
(103.3%)
Stream Add
single-threaded vector
148.0147.1
(99.4%)
Stream Add
multi-threaded vector
180.2179.6
(99.7%)
Stream Triad
single-threaded scalar
99.9104.2
(104.3%)
Stream Triad
multi-threaded scalar
155.5163.8
(105.3%)
Stream Triad
single-threaded vector
115.8116.3
(100.4%)
Stream Triad
multi-threaded vector
145.0141.6
(97.7%)

Conclusion

None of the benchmark results are terribly suprising; the new Mac mini’s processor with a clock speed about 10% faster than the old Mac mini’s processor, and overall the new Mac mini is about 5% faster than the old Mac mini.

Since the new low-end Mac mini is basically the old high-end Mac mini (with a smaller hard drive and a ComboDrive instead of a SuperDrive), this means that there’s not a huge performance difference between the two models. Unless you want the SuperDrive or the extra hard drive space, you’ll be just as happy with the low-end Mac mini as the high-end Mac mini.


HP: How To Tell People They’re Fat Without Really Trying

A number of HP digital cameras now include a feature they call “Slimming.” Short story shorter, it’s a digital effect that slims down the subject of your photos. Now personally, I find this a touch on the offensive side. People have enough body image issues enough without HP telling them can/could/should be slimmer. It’s also worth noting that the two women they use in the demo on their site look healthy and attractive before the effect is applied.

Sane people agree with me that it’s sick/icky/dumb.

Others people, of the less sane variety, seem to think it’s just fine. Maybe they need to give their head a shake.


MacBook Pro Impressions

This summer I made some changes to my lifestyle and career direction and as a consequence of those changes I needed a new computer. My existing machine had been a 1.33 GHz 12″ PowerBook G4 and the one before that had been a 600 MHz 14″ iBook G3. My PowerBook had been a substantial increase in performance at the time, the G4 being a generation newer and more than twice the frequency, it wasn’t hard to choose between it and the iBook of the day for two reasons:

  1. The video card. At the time, the PowerBook was supported for CoreImage and the iBook was not.
  2. The size. My iBook being 14″ and 6 pounds heavy, it was quite the machine to lug around. At the time, I had been lugging it back and forth to class so the prospect of losing a couple inches off the frame and the accompanying 1.4 pounds was quite enticing.

So when I was looking at a new machine I seriously considered the MacBook, and it certainly has its pros and cons. The biggest confusing factor for is that the MacBook and MacBook Pro are just so damn similar, and yet there is a CDN $550 price difference. The CPU, Hard Drive and optical drive are all the same, and the size and weight are no longer that different.

Ultimately when I went to the store with the intent of settling for the MacBook (mostly for monetary reasons) my salesman friend Mark advised me to go for the MacBook Pro. It was a decision I’m glad I let myself be talked into.

Deciding: MacBook vs. MacBook Pro

Things aren’t that different today than they were when I bought my PowerBook. I still need a laptop that’s light, and I still want a laptop that is supported for all the features of Mac OS X (and will be for the next release or two).

Chances are that both the MacBook and MacBook Pro will fit that last requirement. Most of the features (2 GB RAM limit, 2.0 GHz CPU, DVD+/-RW) are the same, in fact the MacBook and MacBook Pro are only separated by a few features besides the video card. The screen, construction and size are the few that matter to me.

To start, the screen on the MacBook has a glossy finish. This style seems to be all the rage these days with proponents constantly telling me that the glossy screens are brighter, clearer and more colour accurate than the satin finish screens that still ship on the MacBook Pros. I tend to agree with people, that the screens are brighter, clearer and more colour accurate as long as there isn’t light shining anywhere near them. I like to work in my living room by big open windows and in coffee shops and outside under trees. In all of these situations there is a lot of light and a lot of glare and a lot of reflection off those glossy screens that renders them near useless, at least for me.

Don’t get me wrong, I hope they make an LCD one day that is brighter, clearer and more colour accurate than the current offerings, I just don’t think we’re there yet.

Another sticking point for me was the fact that the MacBook is made of plastic rather than the metal which encases the MacBook Pro. It’s not that the MacBook feels cheap, indeed compared to most PC laptops it feels incredibly sturdy and incredibly sleek, but compared smooth metal that I’ve become used to with my PowerBook the plastic MacBook just seems, well, cheap. Sure, it’s a metal frame and it’s sturdy as hell, but there is something about the aluminium case of the MacBook pro that just feels better and more professional.

The third and probably most important thing to me when considering laptops is the size. Not to put to fine a point on it, but I like my portable computers to be portable. This is the reason I’ve never seriously considered a PC laptop. Sure, you can get them small, but they often lack features that I want or need. Apple manages to make laptops that are small without sacrificing all that much in the way of features, and I like that.

My 12″ PowerBook was near the perfect size: 4.6 pounds light and fit into a shoulder bag with out much fuss. Today things are a little different in that I don’t need a machine to be quite that small anymore, but more to the point the MacBook doesn’t offer as significant a drop in size and weight over the MacBook Pro anymore: just under half a pound (vs. almost a pound before) and only an inch less wide and deep (vs. a couple inches narrower before). So while the MacBook Pro is a tad smaller and lighter, it isn’t the major selling point that it once was.

The one feature that really sets the MacBook and MacBook pro apart but isn’t of major concern to me, is the video card. The MacBook features an integrated Intel video card with shared memory while the MacBook Pro features a powerful ATI standalone card**.

The integrated card in the MacBook is a fairly powerful one (on paper one of the more powerful integrated cards ever released) but it’s still integrated and while I realistically I probably wouldn’t be using the ATI card to its full potential* it’s more likely that the ATI card will be supported for all the shiny new whiz bang features Apple will be introducing into Mac OS X now (such as CoreImage in Tiger) and in the future (such as CoreAnimation in the upcoming Leopard).

But, even knowing all this, I had still gone into the store with the intention of leaving with a black MacBook. I was going to settle, and that wasn’t sitting well with me. Luckily, Mark is a pretty good salesman and even more luckily for me he had a deal that I couldn’t resist. So instead of leaving the store with a shiny new black MacBook I left with a 2.0 GHz MacBook Pro with a 100 GB hard drive and 1 GB of RAM (which I upgraded to 2 GB almost right away, but that’s another article).

Aesthetics

The appearance of the MacBook Pro is pretty much exactly the same as the previous generation of PowerBooks. Aluminum case, rounded edges, ports on either side. Compared to my 12″ PowerBook though, it feels much more solid. I am not sure exactly what it is but as I mentioned previously this machine just feels like it’s put together better. There is little or no flex in the case when I pick it up and the aluminum has a smooth, sleek feel to it that somehow my PowerBook lacked (or maybe it was just the 18 months of my hands resting on the PowerBook).

Aside from how it’s put together though, the machine just looks cool. Everything from the way the Apple lights up on the back (which is more evenly lit than my PowerBook was) to the backlit keyboard (which looks a hell of a lot cooler in practice in a dark room than it ever did on a showroom floor). Indeed, the MacBook Pro is every bit the head turner that the PowerBooks were before it.

On the sides of the machine, as per usual, is where the ports are. The mere fact that the ports are on the side is one of the many little things that has always had me convinced that Apple thinks a lot harder about what they’re doing then most other manufacturers. Putting the ports on the side instead of the back makes them easier to access and simply makes more sense than having them in the back as the Titanium PowerBooks did and many PC laptops still do today. Many PC manufacturers have started putting ports on the side as well, but they generally speaking still have important ports like the power connector at the back which I find annoying and inconvenient.

The MacBook Pro 15″ lost the FireWire 800 featured on the 15″ PowerBook, and I’m sure irked some people but I haven’t missed it since neither my 12″ PowerBook nor my Windows machine has one either. If I had to complain about the ports I might say that this machine could do with a third or even fourth USB port, but it’s not a major issue.

Really though, the port worth talking about is the MagSafe power connector. I’ve only had this machine for a few weeks but the MagSafe connector has saved my ass no fewer than three times. I know you’re probably thinking “what the hell kind of moron is this guy?” Well, I’m the kind who just moved into a new space and has a mess of cables that get caught on chair wheels, feet, and boxes falling over. My PowerBook had sustained damage to the case where someone tripping on a cable brought the whole thing down off the surface it was sitting on so to say I appreciate the MagSafe connector is a bit of an understatement. Realistically it is an incredily simple little feature but it’s just one more thing that highlights just how much the hardware design team at Apple pays attention.

The 15″ screen on on my MacBook Pro is bright and sharp, and while I do miss the size of my 12″ PowerBook I must confess that I’m now unsure how exactly I got by with only a 1024 x 768 pixel display for the last five years (iBook and PowerBook). The bigger, wider display has increased my productivity in the last two months by a factor of about a million. Well, it’s not like I can really tell.

The keyboard is relatively unchanged from my PowerBook (or even the iBook before it) except that it’s backlit and it feels a little tighter. I haven’t had a chance to play with the new style keyboard on the MacBook, but I am accustomed to this one so to me it’s no big deal.

I will say though that the backlit keyboard, which I’d never before seen in action anywhere other than a showroom floor, is incredibly cool, and the photocell in the machine responds reasonably well when the room gets dark in automatically turning on the backlighting. My only complaints about it are that when the backlighting comes on I prefer it to be on 100% all the time rather than the variable rate that the computer calculates, and that sometimes the photocell gets a little bit confused in low light situations and it ends up turning the backlighting on and off and on again (and for those of you who are about to tell me, I already know about the keyboard buttons that control it).

In short, the outside of this is a beautiful and incredibly functional computer which boasts a few very nice feature updates over its predecessors (as in both the PowerBooks that came before it and my previous machines).

Performance

The Core Duo processor that powers the MacBook Pro represents a substantial increase in performance over the G4 in my PowerBook, but unless you’re living under a rock (or in denial) you already knew that. But what would an article on Geek Patrol be if I didn’t use it to pimp out Geekbench?

I Geekbenched my PowerBook G4 and my MacBook Pro and also found a G3 machine equivalent to my iBook (which yes, I still own but died just over a year ago) in the Geekbench result browser***.

Core Duo vs. G4 vs. G3 performance graph. Click image for bigger

When I upgraded from my iBook to my PowerBook, you can see that represented just over a 100% increase in performance. At the time it was great, but in retrospect the three year gap between my iBook and PowerBook should have represented a much greater increase in performance. Doubling the performance is a lot, but not when you consider the amount of time past and the fact that the PowerBook was only about double the frequency. For contrast, there was only 18 months between my PowerBook and my MacBook Pro and that upgrade represented a nearly 200% increase in performance with only a (roughly) 50% increase in CPU frequency.

The only thing holding this machine back now is that some of the apps I use all the time (ie: Adobe Photoshop) still aren’t universal and likely won’t be for some time, but even that is not as big a concern as it could be.

Rosetta performance graph. Click image for bigger

Even running non-native code, the Core Duo performs almost 50% better than my G4 did, and scores just under our 1.6 GHz G5 baseline result. In other words, I’m still seeing a respectable performance gain over my old machine and that ain’t bad.

Issues: Known, silly and Otherwise

So far my MacBook Pro has been basically issue free. Some programs run slow but I am sure that has more to do with those programs (ie: the non-universal ones) than anything else because most programs run blazingly fast.

It does get hot but this honestly isn’t surprising to me. This is my third Apple laptop and the third one that gets hot enough that it isn’t comfortable to use directly on my lap when it really gets going. The fan controller does seem to work a little more efficiently than it did in either of my previous machines; where the fans in my iBook and PowerBook rarely came on and when they did they did at full tilt, the fans in my MacBook Pro come on relatively often but at reduced speeds cycling up and down again as needed. As I understand it this is how my PowerBook was supposed to work but never did (even after the motherboard was replaced because the first fan controller died).

There is one issue that has cropped up which is purely cosmetic: the paint directly below the keyboard can be easily scratched off and is now a touch patchy. Now, I buy computers to use and not to look at so I personally don’t really care about paint scratching there. I will probably scratch the remaineder of the paint off that thin little strip and it should end up looking pretty much the same as it did before except a little bit shinier, in contrast to the slick matte finish of the rest of the machine that we’ve all come to know and love.

The area beneath the keyboard on my MacBook Pro with the spots where the paint scratched off highlighted. Click image for bigger

The last, and biggest, issue I have with the MacBook Pro is the lack of a dial up modem. Now, this isn’t a huge issue really because I spent most of my time in Vancouver and in Vancouver I have high speed internet almost everywhere I go. I do spend time in places other than the city though, and some of those places still only have dial up internet.

I know you can get an external USB modem for $60 CDN, but it seems somehow premature for a modem to not be included in a laptop. Say what you will about desktop machines, but portable machines are meant to go places and not everywhere has the best access to the internet. Feel free to dredge up arguments about the iMac and the decline of the floppy disk, but in my mind there is a difference between an antiquated storage medium that is barely big enough to hold the files that are being passed around and the antiquated networking method that’s still the only way to get online in the boondocks some of us have to visit.

Final Thought

I’ve been using this machine for over a month full time now and so far I freaking love it. Say what you will about “Rev A” hardware from Apple, but despite the very minor problems mentioned above this machine has been rock solid for me in basically every way.

My only concern really is that Apple might replace the Core Duo CPU with a Merom Core 2 Duo CPU sooner than later and that might sting a little bit; the performance difference aside, the jump to a fully 64 bit CPU would be a nice addition to the other features that make this machine future proof. Realistically I doubt it’ll be an issue though, as I’m sure this machine will last me well into the future. In fact I bet it will last me longer than the 18 months my PowerBook did.

In closing, the 15″ MacBook Pro is a fantastic computer and despite the threat of upcoming changes and the fact that I spent more than I originally intended to, it is a purchase I don’t think I’ll ever regret.

Notes

* I have a desktop PC I use for gaming and such.

** Aside: since ATI and AMD recently merged like two hippos in heat, will we continue to see ATI cards in Intel powered Macs? I’ve no idea, but it’s worth thinking about.

*** The iMac G3 I found isn’t exactly the same machine as my iBook G3 but it is the same CPU, the same bus speed and almost the same amount of RAM (my iBook had 640 MB vs. this iMac with 512 MB, but either way Geekbench only uses 100 MB). In other words, it’s not the same machine but it’s close enough for the purposes of this article.

The Geekbench results I used for this article can be found here:


Showtime: New iTunes, iPods

Unless you live under a rock (or don’t live on the internet as some of us do) you know that Apple held an event today and released a bunch of new toys to we, the seething roaring shouting masses. Here is a quick rundown of what’s come out.

iTunes 7

Version 7 of Apples flagship iApp includes a bunch of new whiz bang features, most of which are pretty cool. Here are my impressions in no particular order.

iTunes 7

The updated iTunes icon.

  • The icon: It’s blue again. Not a huge deal, but as far as I remember this is the first time there has been a major iTunes release without a completely new colour (and no, iTunes 6 was not a major release).

  • Cover art: iTunes will now go through your library and (as long as you have an iTunes Store account and all your ID3 tags are in good order) import cover art for your songs.

    My results with this have been hit or miss so far. It seems to take a long time and even then it doesn’t get everything. For now I’ll stick to iTunes Art Importer on my PC and just wing it on the Mac.

  • CoverFlow: Nifty new way to browse your music by its cover art. Cool looking if a bit slow loading some of the art.

  • Gapless playback: All I can say is that it’s about fucking time.

  • Separate libraries: You can have different libraries on different hard drives and devices. Kind of like vaults in Aperture.

    I haven’t tried it yet, but here’s hoping that unlike Aperture vaults this works over a network.

  • Higher resolution videos: Videos are now encoded at 640 x 480 instead of 320 x 240. Cool, and leads into the next point:

  • Movies: You can buy movies on the iTunes Music Store now.

    Sounds cool, but I’ll likely stick to DVDs. Movies/Videos are now encoded at 4x the resolution that they were previously though, which makes it such that I might actually look at it as an option.

  • iPod games: If I had an iPod that could take advantage this might be cool if I hadn’t already bought all the PopCap games on Steam, and I sincerely doubt the Steam versions are going to get me licenses for my iPod anytime this eon.

  • iPod syncing across multiple computers: See comment regarding “Gapless Playback.”

  • Playback via multiple speakers: Probably my favourite new feature, but it doesn’t even get a mention on Apples iTunes Page.

    You know the little drop down menu you can use to select whether you want to use local speakers or AirTunes speakers? Now you can select both and it plays back in sync through both. It’s so very cool. Now I can fill my house with music. Imagine this feature for parties! You can have music blaring everywhere all at the same time!

    I haven’t had a chance to see if this works for multiple AirTunes set ups yet (because I only have one AirPort Express) but if I find out that it does I’ll likely get another AirPort Express just because of it.

  • New user interface: The iTunes 7 UI seems to share more in common with Apple Pro Apps rather than the iApps. New buttons scroll bars and library organization are all well and good (I don’t mind it but I’m sure some of you do), but I wonder how long until we get a consistent interface across the iLife suite? That’d be super.

    There is also a new list view for music which groups music with its album art which I quite like.

iPods

We also were treated to new iPods today, too.

Full sized

full sized iPods

The full-sized iPod now comes in at 30 GB or 80 GB, with the ability to play the new iPod games, gapless playback support, and longer battery life (14 hours on the 30 GB, 20 hours on the 80 GB). No word yet on whether older 5G iPods are going to be updated for the higher resolution videos or the games. The 30 GB model comes in at $250 USD ($300 CDN) while the 80 GB model is priced at $349 USD ($399 CDN).

nano is the new mini

The iPod nano is now in its second generation and now comes in three sizes: 2 GB, 4 GB and 8 GB. The case has also seen a significant change, now made from aluminum and resembling a smaller, thinner iPod mini and sporting different colours. The 2 GB model is priced at $149 USD ($169 CDN) and comes in silver, the 4 GB model is priced at $199 USD ($229 CDN) and comes in silver, green, blue and pink, and the new top end 8 GB model is priced at $249 USD ($299 CDN) and comes only in black.

The new nano also comes with an increased battery life, up to 24 hours (which is an increase from the previous generations 14 hours or even the mini which had 18 hours).

iPod nano

I was kind of surprised to see the nano looking so much like the mini did, but at the same time it isn’t something I’m going to complain about. Apple’s not really known for regressing in terms of industrial design, but I always preferred the look of the mini to that of the nano.

shuffle

iPod shuffle

The new iPod shuffle is amazing. It now comes in one model: 1 GB. It also sports a shiny, tiny, new silver case with a built in clip for your pocket, belt or whatnot. it also boasts a 12 hour battery which isn’t that long until you consider the size of the unit which is as I previously mentioned, shiny. In all seriousness, the new iPod shuffle is the most impressive of the lot and comes in at $79 USD ($89 CDN).

iTV

Last but certainly not least Apple also announced a new set top media playing box so you can watch or listen to your iTunes content on a television set. There aren’t a lot of details yet, but it outputs video via HDMI or Component (RGB), audio by optical or RCA, can connect to a computer via USB2, ethernet or WiFi and will be priced at $299 USD.

I need to see more details before I make a final judgement about this one. I’d like to know what powers the thing and how big the HDD/active memory (which I’m sure it’ll need for caching) is and so forth. Honestly my first impression is a little bit disappointed though, because in order to use the thing I’d have to get a new television or a component switch box, neither of which is an ideal. I know the video resolution is good, but it’s be nice if the thing had S-Video or RCA video too, since most TVs have more of those than anything else.

Final Thought

All in all it’s been a pretty good day for Apple. Lots of new products and lots of new press. I’m sure they’re going to do well with movies sold via iTunes and I know they’re going to sell a lot of new iPods (I’m going to grab a 4 GB nano. I’m still a little annoyed that movies can’t be burned to a playable disc but I’m sure that has more to do with Apple toeing the MPAA line than anything else.


TextMate and Markdown

While we’re fond of BBEdit here at Geek Patrol, over the last week I’ve started using TextMate for editing Markdown. While BBEdit treats Markdown as just another plain-text format, TextMate provides all sorts of handy shortcuts and utilities for Markdown (as well as having Markdown syntax highlighting built-in). My favourite utility so far is the “Lookup Word on Google and Link”, which looks up the current word (or phrase) on Google and generates a link for it. Very handy!

It turns out I’m not the only person who feels this way; Joel Spolsky (of Joel on Software fame, also likes editing Markdown with TextMate:

I was trying to find appropriate software that I could use to compose long articles that felt smooth on a Mac, that generated extremely clean HTML, and that generated curly quotes (“”) which I’ve grown fond of, especially for longer articles.

The combination I found that made me happiest was TextMate in Markdown mode. It was a surprisingly good experience. TextMate is an “emacs inspired” editor for the Mac, with tons of build-in stuff for editing different types of text files that they call Bundles.

Will TextMate replace BBEdit for me? I’m not sure. I haven’t had a chance to really put TextMate through its paces and find out. I’m cautiously optimistic, though, that I might’ve found an editor that’s more flexible (and more powerful) than BBEdit.


iMac Core Duo and Core 2 Duo Comparison

Apple updated the iMacs last week; among other improvements, Apple moved the iMac from the Core Duo to the Core 2 Duo. We’ve received some Geekbench results for the new iMacs, so I thought I’d take a quick look and see how the new Core 2 Duo iMac compares to the old Core Duo iMac when it comes to performance.

If you’re curious how older iMacs perform (from the G3 to the G5 to the Core Duo), you might want to take a look at our iMac Comparison that we published back in March.

Setup

Here is the configuration of our two test machines:

  • iMac (Early 2006)
    • Intel Core Duo @ 2.0GHz
    • 1024 MB RAM
    • Mac OS X 10.4.7 (Build 8J2135)
    • Geekbench 2006 (Build 200)
  • iMac (Mid 2006)
    • Intel Core 2 Duo @ 2.0GHz
    • 1024 MB RAM
    • Mac OS X 10.4.7 (Build 8K1106)
    • Geekbench 2006 (Build 200)

As always, we’re using the baseline score (not the raw scores) from each benchmark. We’ve also computed the Core 2 Duo’s score as a percentage of the Core Duo’s score (to make the comparison easier). For all scores, higher is better.

Results

Overall Score

Core Duo Core 2 Duo
Overall Score 186.4 204.9
(109.9%)

Integer Performance

Benchmark Core Duo Core 2 Duo
Emulate 6502
single-threaded scalar
91.7121.9
(132.9%)
Emulate 6502
multi-threaded scalar
182.7242.9
(133.0%)
Blowfish
single-threaded scalar
162.6174.7
(107.4%)
Blowfish
multi-threaded scalar
324.0348.8
(107.7%)
bzip2 Compress
single-threaded scalar
148.0168.2
(113.6%)
bzip2 Compress
multi-threaded scalar
291.8333.8
(114.4%)
bzip2 Decompress
single-threaded scalar
135.5212.7
(157.0%)
bzip2 Decompress
multi-threaded scalar
275.4435.7
(158.2%)

Floating Point Performance

Benchmark Core Duo Core 2 Duo
Mandelbrot
single-threaded scalar
141.4134.9
(95.4%)
Mandelbrot
multi-threaded scalar
281.9269.4
(95.6%)
Dot Product
single-threaded scalar
306.1272.7
(89.1%)
Dot Product
multi-threaded scalar
520.3397.8
(76.5%)
Dot Product
single-threaded vector
104.3113.3
(108.6%)
Dot Product
multi-threaded vector
200.0218.3
(109.2%)
JPEG Compress
single-threaded scalar
110.2121.0
(109.8%)
JPEG Compress
multi-threaded scalar
220.1241.0
(109.5%)
JPEG Decompress
single-threaded scalar
101.7115.2
(113.3%)
JPEG Decompress
multi-threaded scalar
199.1227.1
(114.1%)

Memory Performance

Benchmark Core Duo Core 2 Duo
Read Sequential
single-threaded scalar
219.2312.4
(142.5%)
Read Sequential
multi-threaded scalar
187.4231.9
(123.7%)
Write Sequential
single-threaded scalar
223.4419.3
(187.7%)
Write Sequential
multi-threaded scalar
222.9270.8
(121.5%)
Stdlib Allocate
single-threaded scalar
193.9204.1
(105.3%)
Stdlib Allocate
multi-threaded scalar
47.067.3
(143.2%)
Stdlib Write
single-threaded scalar
121.2104.4
(86.1%)
Stdlib Write
multi-threaded scalar
135.6116.8
(86.1%)
Stdlib Copy
single-threaded scalar
245.2218.9
(89.3%)
Stdlib Copy
multi-threaded scalar
213.3254.7
(119.4%)

Stream Performance

Benchmark Core Duo Core 2 Duo
Stream Copy
single-threaded scalar
167.8175.6
(104.6%)
Stream Copy
multi-threaded scalar
177.4172.3
(97.1%)
Stream Copy
single-threaded vector
157.9158.6
(100.4%)
Stream Copy
multi-threaded vector
156.8159.6
(101.8%)
Stream Scale
single-threaded scalar
172.6179.7
(104.1%)
Stream Scale
multi-threaded scalar
184.4183.3
(99.4%)
Stream Scale
single-threaded vector
152.8158.5
(103.7%)
Stream Scale
multi-threaded vector
157.7158.6
(100.6%)
Stream Add
single-threaded scalar
149.2154.9
(103.8%)
Stream Add
multi-threaded scalar
176.1185.0
(105.1%)
Stream Add
single-threaded vector
162.5172.7
(106.3%)
Stream Add
multi-threaded vector
178.9180.0
(100.6%)
Stream Triad
single-threaded scalar
145.1157.6
(108.6%)
Stream Triad
multi-threaded scalar
180.2185.9
(103.2%)
Stream Triad
single-threaded vector
135.5138.9
(102.5%)
Stream Triad
multi-threaded vector
141.3144.9
(102.5%)

Conclusion

As you can see, the Core 2 Duo is 10% faster than the Core Duo. That might not seem terribly impressive, except that the Core 2 Duo is running at the same clock speed as the Core Duo, and the Core 2 Duo iMac is the low-end iMac, while the Core Duo iMac was the high-end iMac. Plus, the Core 2 Duo can run 32- and 64-bit programs, while the Core Duo can only run 32-bit programs.

If you’ve already got an Intel-based Mac, upgrading to a Core 2 Duo doesn’t make a lot of sense (unless you need to run 64-bit programs). If you’re upgrading from any other iMac, the Core 2 Duo iMac shouldn’t disappoint when it comes to raw performance.