The Mage wrote:
> 
> You can notice a 5% improvement, eh? Okay, you've obviously got a way 
> better sense of time than I do... I had no idea I was talking to such a 
> finely tuned specimen! :)

I said I *measured* a 5% improvement...I also got a more stable system, 
which I REALLY noticed, particularly when spending 6-14 hours of a day 
on rendering stuff. When you're rendering 250 frames of animation and 
you get a 5% savings on each frame, that doesn't add up nearly as much 
as the savings from not having it blow up on frame 152 at 3 AM.

>>If you have a system that will run within the interleaved memory all
>>the time, you will notice the difference.
> 
> 
> I did. And I had interleaved memory to spare. And I didn't notice it. I 
> was especially looking for it in web-browsing, and I turned off the 
> browser's disk cache to isolate the difference. There was none. The 
> other apps were just generally the same. Like I said, maybe you've got 
> a better sensor, after all we are human not machines made to order, and 
> 'noticeable' is obviously a very subjective question.

Web browsing is not, generally speaking, memory-constrained. You need to 
do things like manipulate images in Photoshop or a 3D modelling 
application (though they're largely processor constrained, they also 
depend on moving about large chunks of memory at once).

Caching in web browsers is designed to evade the bandwith constraints 
imposed by your network connection, so turning that off likely just 
swamped the diference memory makes. Boosting the memory cache would have 
been a better test.

It's entirely likely that the applications you're using are not memory 
constrained, so a 5% improvent in something the app spends 15% of it's 
time doing is going to be quite unnoticeable.

This is, in a large part, the genesis of the old adage "Liars benchmark 
and benchmarks lie"...you need to be careful to measure your systems so 
that you are, in fact, measuring what you wish to measure ;-)

-- 
Bruce Johnson
University of Arizona
College of Pharmacy
Information Technology Group

Institutions do not have opinions, merely customs




-- 
Unsupported OS X is sponsored by <http://lowendmac.com/>

      Support Low End Mac <http://lowendmac.com/lists/support.html>

Unsupported OS X list info <http://lowendmac.com/lists/unsupported.html>
  --> AOL users, remove "mailto:";
Send list messages to:     <mailto:[EMAIL PROTECTED]>
To unsubscribe, email:     <mailto:[EMAIL PROTECTED]>
For digest mode, email:    <mailto:[EMAIL PROTECTED]>
Subscription questions:    <mailto:[EMAIL PROTECTED]>
Archive <http://www.mail-archive.com/unsupportedosx%40mail.maclaunch.com/>

Using a Mac? Free email & more at Applelinks! http://www.applelinks.com

Reply via email to