> -----Original Message-----
> From: Dave l [mailto:[EMAIL PROTECTED]
> Sent: Monday, April 16, 2007 3:12 AM
> To: CF-Community
> Subject: Re: Looks like MS is trying to shove Vista down consumer's
> throats
>
> "PhotoShop performance really ends up as a wash in my experience.  Some
> things are faster on one platform, some on another - but nothing that
> you'd really notice"
> 
> Well Jim you havent experience photoshop cs3 ub on an intel mac then I
> guess.
> Average startup time for ps on a pc or cs2 on a mac was 40-70 seconds,
> my mbp opens up ps cs3 in around 6 seconds, I think that is more than a
> little bit noticeable. And all the functions render that much faster as
> well, granted its still in beta and might slow down a little bit. Now
> you put that on a new 8 core mp and lord have mercy!

You seem to be mixing comparisons again.  You can't compare CS2 (which runs
terribly on Intel Macs anyway due to Rosetta) on PC to CS3 on Mac.

Tests show no difference in opening speed for CS3 between OS X and Win XP on
the same MacBook, for example (which is actually a little surprising since
the Windows drivers provided with BootCamp are pretty poor, especially disk
performance and startup time should be disk constrained).

Here are some tests (using similar MacBooks running Windows XPsp2 with Apple
BootCamp drivers):

http://hansv.com/cs3/

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

You can see that in the first of these tests there is a marked difference
(about 50% in favor of Mac OS) in actual benchmark time for the first run of
that benchmark.  First runs are, by convention, discarded for this very
reason - things like boot status (starting the benchmark before the OS is
really completed booting), disk caching, first-run virus checking, etc can
all dramatically affect performance on first runs but never again.

The subsequent run is much more even (running about 6% faster on Mac OS).
OS X seems to lose about 45% of its speed but I doubt that's linear.  ;^)
Windows seems to provide pretty even performance across the tests.  It's a
shame because another 10 or 15 runs, averaged, would have given us a much
better picture.

The other test, using a different benchmark and (I think much more
importantly) more RAM, shows that XP beat OS X in performance by over 25%.
I would bet (an educated guess, nothing more) that this test simply shows
that XP makes better use of Large amounts of RAM better than OS X (although
filters in and of themselves tend to be more CPU constrained than RAM
constrained that data has to get to the CPU).  I would guess that the
performance difference would disappear (or reverse) at 1 Gig, but of course
I'm not sure (but OS X does tend to use smaller amounts of RAM more
effectively than XP).

It's impossible to come to a conclusion from these two tests.  But I still
think that actual usage between the two platforms on the same hardware would
be a wash.  Computers spend much more time waiting for us than we do for
them.

Of course some people would argue that similarly configured Macs and PCs are
NOT equal simply because the Mac retains a price-premium.  They would rather
test similarly _priced_ machines, which, at this point tends to favor PCs
since at any price point you can almost definitely double the RAM or up the
proc speed on the PC.  Others would argue that the quality of Windows
drivers provided in Bootcamp favor the Mac - the "homecourt" advantage and
all - and would still prefer similarly configured machine optimized for each
OS.

But the bottom line is the same as I've always said: you'll be more
productive using what you're more comfortable using.  A Mac user will be
more productive on a slow Mac than on fast PC and vice versa.  Benchmarks
test the speed of the machine, not the user.  To my knowledge nobody has
done a task-based analysis between CS3 on Windows and CS3 on the Mac and I
doubt anybody ever will - it really doesn't make much sense.

Jim Davis


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~|
Create Web Applications With ColdFusion MX7 & Flex 2. 
Build powerful, scalable RIAs. Free Trial
http://www.adobe.com/products/coldfusion/flex2/?sdid=RVJS 

Archive: 
http://www.houseoffusion.com/groups/CF-Community/message.cfm/messageid:232613
Subscription: http://www.houseoffusion.com/groups/CF-Community/subscribe.cfm
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.5

Reply via email to