This test isn't as accurate as it could be since so much is dependent on Photoshop preference settings. I'll explain a little later, but first, the test results.
PowerMac G5 Dual 1.8 GHz, 3GB RAM, Mac OSX 10.4.7, 70GB free HD space:
3m 46s 17ms - First test (under my normal PS settings - 30 History States, 30% Memory Usage* - 884 MB)
2m 59s 50ms - Second test (30 History States, 90% Memory Usage - 2652 MB)
1m 28s 85ms - Third test (1 History State, 90% Memory Usage - 2652 MB)
As you can see, depending on how you've configured your PS settings, the results are wildly different. In fact, this test doesn't seem to test anything of discernible value, because it's sooo disk intensive the results skew strictly processor/RAM intensive tests between different systems. The fact that the end file (532 MB) is 15,512 x 12,008 pixels is utterly ridiculous. How many people build 300dpi poster-sized advertisements? This test is great if you're in that niche industry, but it does little to truly test a wide range of systems. An end image of 11x7 inches at 300 dpi should be more than sufficient.
The final disk cache size is slightly under 6GB. Way too disk intense for most systems that have 2GB or under of RAM, like is the case with many older PowerMac systems and PowerBooks.
So I ran the test again, but this time, disabled the 300% image scaling.
0m 17s 63ms - Fourth test (1 History State, 90% Memory Usage - 2652 MB)
Conclusion: You can come to your own conclusions regarding my test. For the person running the Mac Pro test, I would hold judgement until he upgrades the RAM to a production-ready amount (say 4GB) and he checks his PS preferences.
* I run many different apps at the same time as PS, mainly After Effects and FCP, so keeping the memory usage for PS low helps keep my system from slowing down. A setting of 90% is not practical, unless you're strictly running PS and no other program, like Distiller, Quark or InDesign or Acrobat, which in a production environment, could be unproductive
PowerMac G5 Dual 1.8 GHz, 3GB RAM, Mac OSX 10.4.7, 70GB free HD space:
3m 46s 17ms - First test (under my normal PS settings - 30 History States, 30% Memory Usage* - 884 MB)
2m 59s 50ms - Second test (30 History States, 90% Memory Usage - 2652 MB)
1m 28s 85ms - Third test (1 History State, 90% Memory Usage - 2652 MB)
As you can see, depending on how you've configured your PS settings, the results are wildly different. In fact, this test doesn't seem to test anything of discernible value, because it's sooo disk intensive the results skew strictly processor/RAM intensive tests between different systems. The fact that the end file (532 MB) is 15,512 x 12,008 pixels is utterly ridiculous. How many people build 300dpi poster-sized advertisements? This test is great if you're in that niche industry, but it does little to truly test a wide range of systems. An end image of 11x7 inches at 300 dpi should be more than sufficient.
The final disk cache size is slightly under 6GB. Way too disk intense for most systems that have 2GB or under of RAM, like is the case with many older PowerMac systems and PowerBooks.
So I ran the test again, but this time, disabled the 300% image scaling.
0m 17s 63ms - Fourth test (1 History State, 90% Memory Usage - 2652 MB)
Conclusion: You can come to your own conclusions regarding my test. For the person running the Mac Pro test, I would hold judgement until he upgrades the RAM to a production-ready amount (say 4GB) and he checks his PS preferences.
* I run many different apps at the same time as PS, mainly After Effects and FCP, so keeping the memory usage for PS low helps keep my system from slowing down. A setting of 90% is not practical, unless you're strictly running PS and no other program, like Distiller, Quark or InDesign or Acrobat, which in a production environment, could be unproductive
