 |
 |
Worthwhile Free Benchmark App for OS X 10.4.11?
|
|
|
|
|
Administrator  Join Date: Apr 2001
Location: San Antonio TX USA
Status:
Offline
|
|
I have tried XBench, and to quote mduell, it is "poo." A measurement tool should produce similar measurements from one run to the next, and XBench does not, not matter how careful you are to make the conditions identical.
So now I have 4GB of RAM sitting there, waiting to go into my aluminum 2.4GHz 20" iMac, and I want to be able to quantify how much difference there is between the 2GB in it right now and the 4GB I am about to put into it. My scientific mind wants to know, but my grad-student wallet says "it has to be free!" 
|
|
Glenn -----
OTR/L, MOT, Tx
|
| |
|
|
|
|
|
|
|
Posting Junkie
Join Date: Oct 2005
Location: Houston, TX
Status:
Offline
|
|
There are a million benchmarks you could run that would or would not show an improvement, and every single one of them would be meaningless.
Benchmark whatever it is that you do. That is the only relevant gain.
|
|
|
| |
|
|
|
|
|
|
|
Administrator Join Date: Apr 2001
Location: San Antonio TX USA
Status:
Offline
|
|
Good point; the most successful computer benchmarks I've ever heard of were suites of simulations, such as PC Magazine's suite. But how does one benchmark loading a page in Firefox or searching a Word document? Stopwatch? Some app?
|
|
Glenn -----
OTR/L, MOT, Tx
|
| |
|
|
|
|
|
|
|
Posting Junkie
Join Date: Oct 2005
Location: Houston, TX
Status:
Offline
|
|
They probably have an app with a timer to load/render/search really big files, or doing it a thousand times in a loop.
But what do you do with your computer? Benchmark that.
If the answer is Word and Firefox, your benchmark can be number of page outs. 
|
|
|
| |
|
|
|
|
|
|
|
Administrator Join Date: Apr 2001
Location: San Antonio TX USA
Status:
Offline
|
|
Cool, thanks. Pageouts I can find! That makes it much simpler to tally so I can move on and actually install the RAM!
|
|
Glenn -----
OTR/L, MOT, Tx
|
| |
|
|
|
|
|
|
|
Mac Elite
Join Date: Sep 2005
Location: Vancouver, BC
Status:
Offline
|
|
Given that you aren't changing the RAM bandwidth with the upgrade, the only case where adding RAM would increase performance visibly in an application is if you're working with huge files that were already on the cusp, or past the cusp, of hitting swap memory.
:-P
Pageouts is really the most useful thing to look.
|
|
|
| |
|
|
|
|
|
|
|
Administrator Join Date: Apr 2001
Location: San Antonio TX USA
Status:
Offline
|
|
Well I'm getting zero page outs with Firefox, and with some pretty simple but processor (and I thought RAM) intensive Word manipulations. So I'm not upgrading to the most RAM possible for scientific reasons-at least not with what I do normally. That'll be ok.
By the way, I have tens of thousands of page INs. I'm now a bit confused about what a page in is-since I haven't paged anything out, is this an indication of predictive caching or something else?
|
|
Glenn -----
OTR/L, MOT, Tx
|
| |
|
|
|
|
|
|
|
Posting Junkie
Join Date: Oct 2005
Location: Houston, TX
Status:
Offline
|
|
Originally Posted by ghporter 
By the way, I have tens of thousands of page INs. I'm now a bit confused about what a page in is-since I haven't paged anything out, is this an indication of predictive caching or something else?
Paging is separate from caching, but they can work together. Page in means the OS copied some memory to disk, just in case it needs that memory later. Page out means the OS had to blow away something in memory and then pull it back from disk later. Caching is taking often used files from the disk and keeping a copy in RAM. Both strategies can be used independently or together for better performance.
|
|
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
 
|
|
|
Forum Rules
|
|
|
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
|
HTML code is Off
|
|
|
|
|
|
|
|
|
|
|
|
|
Top
