Currently Being ModeratedJun 1, 2013 2:35 AM (in response to hoaxe)
Firstly, it is not recommended nor necessary to use third party memory management applications. Mountain Lion has an autonomous system which is very good at it's job.
In Activity Monitor you can see how much memory is in use. In the screen shot below, the green area of the pie chart is free memory and the blue is inactive memory. Essentially the blue area still holds information from recent usage and therefore can speed up re-use of this data, so a large amount of green "free" memory is actually a waste of capacity.
What is more relevant is the "Page Out" figure as this may indicate that data needs to be written out from RAM to the hard disk. Too little RAM may cause this to happen.
Since the MBA has SSD and this is fast memory too, having 2Gb of RAM is not the limitation it would have been with an older hard disk.
If your computer shows evidence of low memory I would suggest reducing the number of apps running at any one time, although even that is not necessary if the Page Out quantity is low or nil.
Currently Being ModeratedJun 1, 2013 8:34 AM (in response to putnik)
This helps me to understand a bit more, but don't solve my problem.
Is it 10.8.x that chokes my old cmp, can it be solved through feature updates of OSX or is it just time to go for new he?
Thanks for both quick and good answer.
Currently Being ModeratedJun 1, 2013 8:45 AM (in response to hoaxe)
At the moment it is Fiplab that is choking your computer. It is the very act of reading and writing between the RAM and the hard disk that causes the slowness. The app is writing out to disk to "free memory" and make the pie chart nice and green. But much of the same information will then need to be written back to RAM again. You must get rid of Fiplab.
You could try some graphics test programs such as Geekbench, to see if your computer is good or bad compared with others. But I still do fine with a 2009 MBP albeit now with 8Gb of RAM but not the advantage of SSD.
Currently Being ModeratedJun 1, 2013 10:28 AM (in response to putnik)
Sweet! I'll uninstall fiplab asap and try the benchmarking link this weekend. Doh, I thought fiplab would help, not sink the HW. Thanks a lot! I'll reply when I tested your fine tips.
Currently Being ModeratedJun 1, 2013 10:37 AM (in response to hoaxe)
Currently Being ModeratedJun 2, 2013 10:57 AM (in response to putnik)
Currently Being ModeratedJun 2, 2013 11:42 AM (in response to hoaxe)
Is that with or without Fiplab? It would be interesting to see your scores for both cases. Perhaps too to research other people's scores for your model.
Currently Being ModeratedJun 2, 2013 11:11 PM (in response to putnik)
Here's a score with fiplab memory clean:
And here's the score without fiblan memory clean:
And the socre is a bit better without the "memory manager", but it's not enough i think. The test was done without any running application (i killed all nonnessasary) and i did run "purge" before.
Why is your box better and faster? looking at the figures - can it be a cpu issue? Looks like your box has better score on integers and floating points.
Currently Being ModeratedJun 3, 2013 2:05 AM (in response to hoaxe)
My first thought is that running a "purge" is going to clear out the inactive (blue) memory. This then needs to be repopulated, so may actually slow it down temporarily. Also, there are large variations in scores and newer models are blisteringly fast compared to yours and mine. Also, it will be interesting to see what effect 10.8.4 has on your MBA, if any.
Here is a comparison chart I found. I expect they are run with the 64 bit app and may be more accurate.
Currently Being ModeratedJun 8, 2013 11:12 AM (in response to putnik)
I've install the 10.8.4 update, but i can't "feel" or measure any improvment for my MBA, yet. I run a benchmark and i got about the same score, and as you wrote..." Also, it will be interesting to see what effect 10.8.4 has on your MBA, if any". You probably right
Thanks for the support and valuable input.
Currently Being ModeratedJun 8, 2013 12:16 PM (in response to hoaxe)
Your welcome to my input. I think Geekbench scores relate mainly to the CPU chip, but other factors such as RAM also affect the user experience. I'll think about updating when the new MBP comes out. Blow the kid's inheritance