PDA

View Full Version : Adobe Photoshop CS3 Beta SLOWER on OSX compared to XP




brandonh
Jan 15, 2007, 04:29 PM
The test results speak for themselves. It's easy to see that Windows XP performs faster for all but 1 of the tests performed. As a total, with the Mushkin 2x1GB and Kingston 2x2GB memory configurations, the test suite ran 27% and 28%, respectively, faster with Windows XP than OSX. Not only does Windows XP beat OSX in Photoshop CS3 performance, but it demolishes it by a wide margin.

http://www.circuitremix.com/index.php?q=node/9

Anyone have any questions, suggestions or criticisms on the test or methods?



irishgrizzly
Jan 15, 2007, 04:41 PM
That is quite a huge margin! I was expecting a difference of more like 5-10%. Bad news for people trying to justify getting a suite of macs for there design company. Hopefully this gap will shrink before the final release.

psychofreak
Jan 15, 2007, 04:44 PM
Most people have more apps open at the same time, macs are better at this.

PCs get considerably slower very quickly, so wait a year and use the same machines (if they've been used), and the results may be reversed.

Photoshop is much nicer on macs!

clevin
Jan 15, 2007, 04:56 PM
they are running on same machine, so i don't see any hardware issue.

if its slower, its slower, psychofreak might be right about it at some points, altho that shouldn't be a reason to celebrate (u are not waiting for a year to use the app).

Blue Velvet
Jan 15, 2007, 04:59 PM
Bad news for people trying to justify getting a suite of macs for there design company.


It's a good thing that the decisions on running and equipping a design studio are not based on running a filter over an image and timing it with a stopwatch. ;)

irishgrizzly
Jan 15, 2007, 05:03 PM
It's a good thing that the decisions on running and equipping a design studio are not based on running a filter over an image and timing it with a stopwatch. ;)

No, just by blindly trusting in anything apple sends us :D

killmoms
Jan 15, 2007, 05:04 PM
It's a good thing that the decisions on running and equipping a design studio are not based on running a filter over an image and timing it with a stopwatch. ;)

Hear hear. It's about total productivity, not benchmarks.

bousozoku
Jan 15, 2007, 05:16 PM
I'd like to see how they perform with the foreground application's priority set the same as other processes in Windows or set higher in Mac OS X to see how that affects the results.

It's not unusual to see Windows foreground process outperforming the background processes because of this. It's not possible to do this through the GUI on Mac OS X but it can be done.

I wouldn't be surprised that the Windows version of Photoshop CS3 outperforms the Mac OS X version but I'd be surprised if it works as well where it counts.

brandonh
Jan 15, 2007, 05:25 PM
I'd like to see how they perform with the foreground application's priority set the same as other processes in Windows or set higher in Mac OS X to see how that affects the results.

It's not unusual to see Windows foreground process outperforming the background processes because of this. It's not possible to do this through the GUI on Mac OS X but it can be done.\

What do you mean by this?

In Windows: Control Panel -> System -> Advanced tab - > Performance Settings -> Advanced tab

... then changing the options from Programs to Background services?

Veritas&Equitas
Jan 15, 2007, 06:13 PM
Hear hear. It's about total productivity, not benchmarks.
You guys can paint these results any way you want, but in reality, this is not good. Not good at all. If you guys take the blinders off for one second, you might realize that "slower" results for a OS X when compared to XP = bad. Don't try to rationalize the results.

Daveway
Jan 15, 2007, 06:16 PM
I thought there was an article published not long ago, when the beta was first released, that said the mac was actually faster in nearly all operations when running a script.

:confused:

Edit:
http://hansv.com/cs3/

tipdrill407
Jan 15, 2007, 06:27 PM
It's just a Beta everyone... a lot can change when the final version is released.

brandonh
Jan 15, 2007, 06:58 PM
I thought there was an article published not long ago, when the beta was first released, that said the mac was actually faster in nearly all operations when running a script.

:confused:

Edit:
http://hansv.com/cs3/

In that test they had Windows XP on a 9GB partition. I don't know much about their setup or their test suite, but that doesn't leave much for scratch disk space or page file (if necessary).

I'll try that out later tonight though.

Blue Velvet
Jan 16, 2007, 01:10 AM
You guys can paint these results any way you want, but in reality, this is not good. Not good at all. If you guys take the blinders off for one second, you might realize that "slower" results for a OS X when compared to XP = bad. Don't try to rationalize the results.



It's not about rationalising, it's about realising that in terms of print design alone, using a Mac is the defacto standard for good reasons.

We're not going to suddenly get rid of all our G5s and purchase new hardware and software and fonts because we can shave a few seconds off running a filter (something I rarely do anyway).

Besides, if something is taking a long time to chew through something on the Mac, distilling a big PDF for instance, I can still get on and do other things. The context is overall productivity and creativity within an established workflow, something that people in the business well understand, particularly if they're smallish setups and also have to do their own tech support.

So to me, this is more 'by the numbers' thinking, isolated from context. It may come as a surprise to some but having slightly faster software doesn't make you a better designer, production artist, or visual communicator.

bousozoku
Jan 16, 2007, 02:23 AM
What do you mean by this?

In Windows: Control Panel -> System -> Advanced tab - > Performance Settings -> Advanced tab

... then changing the options from Programs to Background services?

Perhaps. I don't have a Windows machine in front of me now.

You guys can paint these results any way you want, but in reality, this is not good. Not good at all. If you guys take the blinders off for one second, you might realize that "slower" results for a OS X when compared to XP = bad. Don't try to rationalize the results.

It might make a difference here and there but it's more important for a rendering farm to have such an advantage than an individual graphic designer.

RacerX
Jan 16, 2007, 04:55 AM
We're not going to suddenly get rid of all our G5s and purchase new hardware and software and fonts because we can shave a few seconds off running a filter (something I rarely do anyway).In fact I know many designers who haven't replaced their G4s yet.

True productivity has never been measured by the speed at which a computer can execute any task... it has always been measured by how well the users of those computers can interact with them to finish the job at hand.

If it was the reverse, then we'd all be buying a bunch of Linux clusters that can execute tasks at speeds unmatched by either Macs or Windows based systems.

In the end, the right set of tools for a job aren't based on horse power alone. I can do tons of things that most computer users can't on systems that range in age from 7 to 12 years old. Why? Because I have the tools needed at hand, they didn't cost me much, and I'm exceptionally productive in the Mac environment.

The fastest Windows system in the world can't offer me that, which is why I've turned down free PCs on numerous occasions. Even completely free (and faster than any Mac I own), a Windows based system isn't worth the space it would take up.


The only time that benchmarks like these are applicable is when all other factors are the same. Seeing as Photoshop on a Windows system is still an inferior product to Photoshop on a Mac, no amount of raw speed increase can ever make up for the impact to a designers productivity.

clevin
Jan 16, 2007, 06:33 AM
well, guys, just remember next time if u find something runs faster on mac, don't bash PC for speed then.:p