Mac Performance: From the G3 to the Xeon
Since we released Geekbench 2006 on July 10th, we’ve received over 4800 Geekbench results; over half of those results are from computers running Mac OS X. I took a look at the results, and found that we’ve received results for almost every computer model Apple’s shipped in the past six years, from the iMac G3 to the Mac Pro.
Since we love us some benchmarks here at Geek Patrol, I thought it’d be interesting to put together an overview of Mac performance over the past six years.
To make things easier, I’ve split the models into four categories: Pro Desktops, Pro Laptops, Consumer Desktops, and Consumer Laptops. I’ve then calculated and graphed the minimum overall score, average overall score, and maximum overall score for each model in each category.
There are a number of models I’ve not been able to include (like the PowerBook G3, the original 17-inch PowerBook G4, and the Xserves) due to a lack of results; any model included in this comparison has at least three results in our result browser.
Pro Desktop
What’s remarkable is that despite all the hubub about Apple failing to ship a Power Mac G5 with a 3GHz processor, Power Mac G5 performance has increased steadily (especially when compared to the Power Mac G4) with each new model.
The Mac Pro didn’t bring a huge performance jump over the Power Mac G5 (unlike like the jump in performance from the Power Mac G4 to the Power Mac G5), but the switch to Intel should ensure the Mac Pro keeps getting faster and faster with each new model.
Pro Laptop
Pro Laptop performance has been virtually stable (or stagnant, if you prefer) since the introduction of the aluminum PowerBooks, which shows just how out-dated the PowerPC G4 was then and now. Now with the switch to Intel, Macs are once again competitive when it comes to laptop performance; performance almost tripled with the switch from the PowerPC G4 to the Intel Core Duo.
Consumer Desktop
iMac performance (unlike PowerBook performace) has increased with each new model. The iMac has had four different processors, and performance has jumped each time the iMac has switched processors. The most impressive switch was the switch from the PowerPC G4 to the PowerPC G5, where performace almost doubled. The switch from the PowerPC G5 to the Intel Core Duo also seems impressive, but a large part of the performace gain came not from switvching architectures, but from switching from a single-core to a dual-core CPU.
Mac mini performance increased more dramatically with the switch to Intel chips, since previous Mac mini models used the G4, not the G5. Now, the Mac mini is an small and elegant machine with lots of performace, especially with the Core Duo as its CPU.
eMac performace, while somewhat lacklustre, did improve with each new model. The eMac is no longer available, though, having been replaced by the educational iMac.
Consumer Laptop
Like the Mac mini, iBook performance jumped dramatically with the switch from the PowerPC G4 to the Intel Core Duo; performance almost tripled (the switch from the PowerPC G3 to the PowerPC G4 was less impressive). MacBook performace is so impressive it’s possible that MacBook sales are cannibalizing MacBook Pro sales.
Final Thoughts
What struck me when looking at the graphs was how much the PowerPC G4 held back the Mac in terms of performance. Even though the G4 was introduced in 1999, Apple was still shipping computers that used the G4 until May 2006, almost seven years later.
Now, with the switch to Intel, Apple won’t be in a similar situation again (or if they are, they’ll be in it with the rest of the industry).
Geekbench 2006 Updated (Build 200)
The latest update to Geekbench 2006 (build 200 for those of you keeping score at home) is now available for download. With this release, Geekbench 2006 now runs natively on 64-bit Windows machines (i.e., it’s a full-blown 64-bit application). Also, Geekbench 2006 for Mac OS X is now built with Xcode 2.4.
Check out the release notes for all the details.
Mac Pro 3GHz Benchmarks
Right after we started receiving results for Mac Pros at 2.66GHz, we’ve started receiving Mac benchmark benchmark results for Mac Pros at 3.0GHz. Since we’ve already compared the mid-range Mac Pro against the top-of-the-line Power Mac G5, we thought we’d compare the top-of-the-line Mac Pro against the same Power Mac G5. Let’s see if the 3.0GHz Mac Pro fares better than the 2.66GHz Mac Pro against a Quad Power Mac G5.
Setup
Here is the configuration of our two test machines:
- Mac Pro
- Intel Xeon 5160 @ 3.0GHz (two dual-core processors)
- 4096MB RAM
- Mac OS X 10.4.7 (Build 8K1079)
- Geekbench 2006 (Build 190)
- Power Mac G5
- PowerPC G5 @ 2.5GHz (two dual-core processors)
- 1024MB RAM
- Mac OS X 10.4.7 (Build 8J135)
- Geekbench 2006 (Build 180)
As per usual, we’re using the baseline scores (not the raw scores) from each benchmark. We’ve also computed the Mac Pro’s score as a percentage of the Power Mac G5’s score. Higher is better.
Results
Overall Score
| Power Mac G5 | Mac Pro | |
|---|---|---|
| Overall Score | 278.3 | 384.1 (138.0%) |
Integer Performance
| Benchmark | Power Mac G5 | Mac Pro |
|---|---|---|
| Emulate 6502 single-threaded scalar |
157.9 | 186.1 (117.9%) |
| Emulate 6502 multi-threaded scalar |
622.4 | 734.7 (118.0%) |
| Blowfish single-threaded scalar |
164.1 | 261.9 (159.6%) |
| Blowfish multi-threaded scalar |
651.5 | 1044.3 (160.3%) |
| bzip2 Compress single-threaded scalar |
157.0 | 250.7 (159.7%) |
| bzip2 Compress multi-threaded scalar |
572.5 | 964.9 (168.5%) |
| bzip2 Decompress single-threaded scalar |
152.8 | 274.4 (179.6%) |
| bzip2 Decompress multi-threaded scalar |
603.0 | 1152.0 (191.0%) |
Floating Point Performance
| Benchmark | Power Mac G5 | Mac Pro |
|---|---|---|
| Mandelbrot single-threaded scalar |
156.5 | 202.4 (129.3%) |
| Mandelbrot multi-threaded scalar |
623.5 | 808.5 (129.7%) |
| Dot Product single-threaded scalar |
143.2 | 408.0 (284.9%) |
| Dot Product multi-threaded scalar |
513.9 | 1214.9 (236.4%) |
| Dot Product single-threaded vector |
152.8 | 171.5 (112.2%) |
| Dot Product multi-threaded vector |
504.7 | 440.3 (87.2%) |
| JPEG Compress single-threaded scalar |
156.7 | 179.6 (114.6%) |
| JPEG Compress multi-threaded scalar |
621.6 | 719.6 (115.8%) |
| JPEG Decompress single-threaded scalar |
160.1 | 172.7 (107.9%) |
| JPEG Decompress multi-threaded scalar |
508.1 | 599.8 (118.0%) |
Memory Performance
| Benchmark | Power Mac G5 | Mac Pro |
|---|---|---|
| Read Sequential single-threaded scalar |
171.2 | 346.0 (202.1%) |
| Read Sequential multi-threaded scalar |
53.4 | 183.1 (342.9%) |
| Write Sequential single-threaded scalar |
196.2 | 556.7 (283.7%) |
| Write Sequential multi-threaded scalar |
185.1 | 272.7 (147.3%) |
| Stdlib Allocate single-threaded scalar |
119.6 | 320.8 (268.2%) |
| Stdlib Allocate multi-threaded scalar |
31.3 | 40.7 (130.0%) |
| Stdlib Write single-threaded scalar |
270.3 | 120.8 (44.7%) |
| Stdlib Write multi-threaded scalar |
367.4 | 206.4 (56.2%) |
| Stdlib Copy single-threaded scalar |
187.6 | 259.4 (138.3%) |
| Stdlib Copy multi-threaded scalar |
259.2 | 421.9 (162.8%) |
Stream Performance
| Benchmark | Power Mac G5 | Mac Pro |
|---|---|---|
| Stream Copy single-threaded scalar |
176.5 | 214.9 (121.8%) |
| Stream Copy multi-threaded scalar |
306.2 | 361.2 (118.0%) |
| Stream Copy single-threaded vector |
190.6 | 210.6 (110.5%) |
| Stream Copy multi-threaded vector |
277.4 | 332.3 (119.8%) |
| Stream Scale single-threaded scalar |
165.7 | 231.8 (139.9%) |
| Stream Scale multi-threaded scalar |
321.5 | 390.2 (121.4%) |
| Stream Scale single-threaded vector |
185.9 | 212.3 (114.2%) |
| Stream Scale multi-threaded vector |
274.9 | 340.2 (123.8%) |
| Stream Add single-threaded scalar |
147.0 | 192.8 (131.2%) |
| Stream Add multi-threaded scalar |
301.9 | 353.6 (117.1%) |
| Stream Add single-threaded vector |
188.4 | 209.9 (111.4%) |
| Stream Add multi-threaded vector |
290.7 | 341.5 (117.5%) |
| Stream Triad single-threaded scalar |
144.4 | 195.3 (135.2%) |
| Stream Triad multi-threaded scalar |
305.4 | 354.9 (116.2%) |
| Stream Triad single-threaded vector |
195.6 | 170.4 (87.1%) |
| Stream Triad multi-threaded vector |
307.2 | 275.7 (89.7%) |
Conclusion
Overall, the 3.0GHz Mac Pro is 35% faster than the Quad Power Mac G5 (and 28% faster than the 2.66GHz Mac Pro); it’s up to 242% faster in particular individual tests (and, unfortunately, up to 100% slower in others). Just like the 2.66GHz Mac Pro, the 3.0GHz Mac Pro excels at integer calulations and floating point calculations. Unlike the 2.66GHz Mac Pro, the 3.0GHz Mac Pro also excels at memory operations, although this might have more to do with the number of DIMMs installed in each machine rather than the processing power of each machine (the 3.0GHz has more DIMMs, and thus has a wider memory bus).
Mac Pro Benchmarks
Even though they were announced only two days ago, we’ve started receiving Geekbench (our Mac benchmark) results for Mac Pros. We thought we’d compare a Mac Pro against a Power Mac G5 (specifically a Quad Power Mac G5) so we could see just how much faster the new Mac Pro really is.
Update: We’ve published more Mac Pro benchmarks, where we compare a 3.0GHz Mac Pro (as opposed to the 2.66GHz Mac Pro in this article) against a Quad Power Mac G5.
Setup
Here is the configuration of our two test machines:
- Mac Pro
- Intel Xeon 5150 @ 2.66GHz (two dual-core processors)
- 1024MB RAM
- Mac OS X 10.4.7 (Build 8K1079)
- Geekbench 2006 (Build 190)
- Power Mac G5
- PowerPC G5 @ 2.5GHz (two dual-core processors)
- 1024MB RAM
- Mac OS X 10.4.7 (Build 8J135)
- Geekbench 2006 (Build 180)
As per usual, we’re using the baseline scores (not the raw scores) from each benchmark. We’ve also computed the Mac Pro’s score as a percentage of the Power Mac G5’s score. Higher is better.
Results
Overall Score
| Power Mac G5 | Mac Pro | |
|---|---|---|
| Overall Score | 278.3 | 297.9 (107.0%) |
Integer Performance
| Benchmark | Power Mac G5 | Mac Pro |
|---|---|---|
| Emulate 6502 single-threaded scalar |
157.9 | 164.8 (104.4%) |
| Emulate 6502 multi-threaded scalar |
622.4 | 648.6 (104.2%) |
| Blowfish single-threaded scalar |
164.1 | 232.6 (141.7%) |
| Blowfish multi-threaded scalar |
651.5 | 924.8 (141.9%) |
| bzip2 Compress single-threaded scalar |
157.0 | 222.5 (141.7%) |
| bzip2 Compress multi-threaded scalar |
572.5 | 859.3 (150.1%) |
| bzip2 Decompress single-threaded scalar |
152.8 | 249.2 (163.1%) |
| bzip2 Decompress multi-threaded scalar |
603.0 | 1047.7 (173.7%) |
Floating Point Performance
| Benchmark | Power Mac G5 | Mac Pro |
|---|---|---|
| Mandelbrot single-threaded scalar |
156.5 | 179.9 (115.0%) |
| Mandelbrot multi-threaded scalar |
623.5 | 716.2 (114.9%) |
| Dot Product single-threaded scalar |
143.2 | 358.6 (250.4%) |
| Dot Product multi-threaded scalar |
513.9 | 990.8 (192.8%) |
| Dot Product single-threaded vector |
152.8 | 144.9 (94.8%) |
| Dot Product multi-threaded vector |
504.7 | 425.4 (84.3%) |
| JPEG Compress single-threaded scalar |
156.7 | 161.6 (103.1%) |
| JPEG Compress multi-threaded scalar |
621.6 | 637.9 (102.6%) |
| JPEG Decompress single-threaded scalar |
160.1 | 154.5 (96.5%) |
| JPEG Decompress multi-threaded scalar |
508.1 | 527.4 (103.8%) |
Memory Performance
| Benchmark | Power Mac G5 | Mac Pro |
|---|---|---|
| Read Sequential single-threaded scalar |
171.2 | 329.1 (192.2%) |
| Read Sequential multi-threaded scalar |
53.4 | 124.0 (232.2%) |
| Write Sequential single-threaded scalar |
196.2 | 550.2 (280.4%) |
| Write Sequential multi-threaded scalar |
185.1 | 120.1 (64.9%) |
| Stdlib Allocate single-threaded scalar |
119.6 | 287.9 (240.7%) |
| Stdlib Allocate multi-threaded scalar |
31.3 | 41.9 (133.9%) |
| Stdlib Write single-threaded scalar |
270.3 | 94.4 (34.9%) |
| Stdlib Write multi-threaded scalar |
367.4 | 92.9 (25.3%) |
| Stdlib Copy single-threaded scalar |
187.6 | 198.4 (105.8%) |
| Stdlib Copy multi-threaded scalar |
259.2 | 198.3 (76.5%) |
Stream Performance
| Benchmark | Power Mac G5 | Mac Pro |
|---|---|---|
| Stream Copy single-threaded scalar |
176.5 | 162.4 (92.0%) |
| Stream Copy multi-threaded scalar |
306.2 | 157.5 (51.4%) |
| Stream Copy single-threaded vector |
190.6 | 152.4 (80.0%) |
| Stream Copy multi-threaded vector |
277.4 | 143.6 (51.8%) |
| Stream Scale single-threaded scalar |
165.7 | 171.9 (103.7%) |
| Stream Scale multi-threaded scalar |
321.5 | 166.9 (51.9%) |
| Stream Scale single-threaded vector |
185.9 | 151.1 (81.3%) |
| Stream Scale multi-threaded vector |
274.9 | 142.4 (51.8%) |
| Stream Add single-threaded scalar |
147.0 | 150.7 (102.5%) |
| Stream Add multi-threaded scalar |
301.9 | 161.0 (53.3%) |
| Stream Add single-threaded vector |
188.4 | 152.4 (80.9%) |
| Stream Add multi-threaded vector |
290.7 | 152.3 (52.4%) |
| Stream Triad single-threaded scalar |
144.4 | 151.0 (104.6%) |
| Stream Triad multi-threaded scalar |
305.4 | 161.6 (52.9%) |
| Stream Triad single-threaded vector |
195.6 | 122.7 (62.7%) |
| Stream Triad multi-threaded vector |
307.2 | 122.4 (39.8%) |
Conclusion
Overall, the Mac Pro is 7% faster than the Power Mac G5. In individual tests, the Mac Pro is between 34.9% and 280.4% of the performance of the Power Mac G5. The Mac Pro outperforms the Power Mac G5 when it comes to integer calculations and scalar floating point calculations, while the Power Mac G5 outperforms the Mac Pro when it comes to vector floating point calculations and standard library memory performance.
Of course, we’ve tested a mid-range Mac Pro ($2500 US) against a top-of-the-line Power Mac G5 ($3000 USD), so even though the Mac Pro is only a little bit faster than the Power Mac G5, it’s significantly cheaper!
Entries