View Full Version : Just put in 8GB RAM, don't notice much difference, URLs seem to load slower?
Jun 15, 2011, 12:38 AM
Hey you guys, after an internal argument on whether it was necessary to buy more RAM for my 2010 Mac Mini, I finally did it. I bought two 4GB Kingston RAM sticks on Friday, and they got here today around 12 :)
I just got around to putting them in, resetting PRAM, and booting up into Mac OS, and everything loaded slower. Is this usual after a first boot after installing new RAM?
Things that were slower:
Boot (45-60 seconds)
What's the deal here?
Thanks in advance for any help, you guys always pull through for me!
Jun 15, 2011, 01:03 AM
First, re-check the RAM specs and be completely sure it's compatible with your computer. Second, make sure it's plugged in right - take it out carefully and put it back in again. (Remember doing this with old video game cartridges?) Next, test your RAM:
http://www.macupdate.com/app/mac/15837/rember (optional GUI)
The most thorough test you can run with memtest is probably excessive. Just run the normal test unless you have several hours to spare.
Jun 15, 2011, 01:18 AM
After a second boot, everything seemed a ton faster. New RAM was detected (just like the first boot), but apps and webpages loaded furiously fast. I think all is well. Should I still have any concerns or run those memory tests?
Safari (with 4 tabs)
Twitter for Mac
and I have 6.18GB of Memory free! :)
Thanks for your quick response!
Jun 15, 2011, 01:35 AM
Nah, don't bother if it seems OK.
I think one reason first-boots take longer isn't just caching, but the system needs to start in a fail-safe mode (so if you lose power it doesn't brick your computer). That's just a guess, I really don't know what I'm talking about.
Jun 15, 2011, 09:44 PM
Rember Test Results
All tests passed!
Total built-in memory: 8 GB
This is the total amount of physical memory that the computer has installed. If this figure is not showing the correct amount of memory there may be a problem with one or more installed DIMMS.
Available memory: 6610 MB
Available memory is the amount of physical memory that is currently not in use by any other processes. All available memory will be used for testing when the "All" option is selected. To increase the amount of available memory, you can restart your computer before testing. If you are familiar with the command line (CLI), you can run memtest (the core of Rember) in single-user mode. See Rember help, or http://www.memtestosx.org for more information.
Requested amount: All MB
The total amount of memory requested for testing by the Rember application. Not all requested memory can be allocated for testing. See information on "Available memory" for more information.
Memory allocated for testing: 6610 MB
This is the total amount of memory that memtest was able to allocate for testing. See "Available memory" section for more information.
Loops selected: 1
Total loops selected by user for testing. All loops should complete when testing is successful. Test failure when the "Continue on Error" preference is selected will cancel tests before this number of loops has been completed. Users can also cancel testing before this number is reached.
Loops completed: 1
Total loops completed by memtest. Note that the Rember is not always able to identify how many loops ran. If there are discrepancies between this and the loops selected, the log should be examined to determine exactly how many loops were performed.
Total execution time: ok
All tests passed! Execution time: 6179 seconds
This is the total amount of time that it took to execute the selected tests. Execution time may vary from system to system, and is provided as a guide for determining how long users can expect tests to run based on the amount of memory installed on the system.
Testing start time: 2011-06-15 13:58:37 -0500
Testing end time: 2011-06-15 15:41:40 -0500
Rember version: 0.3.7b Memtest version: 4.22
Does this and the images provided look correct?
Jun 15, 2011, 10:44 PM
yes it passed and should be good for years to come.
Jun 19, 2011, 03:18 AM
I hope you didn't expect a webpage to load faster with more ram? Even apps usually don't start faster with more ram.
The only reason an app would start faster if when you run out of ram before/during the startup of the app and the machine would start swapping. Adding ram is not as upping the clockspeed of the cpu. The latter always helps, the former helps when you run out of ram.
There is a rather logical explanation ram doesn't make apps start faster: An app is started when it's loaded from HDD to Ram to cache to CPU (there are also prefetch caches and ram caches, but lets keep it simple). Adding ram onlu ups the pool of avialable ram, it does nothing to speed up this basic process.
now if you run out of Ram it becomes a different ballpark. The program with the lowest priority (algorithmically decided by OS) gets it's ram offloaded to the HDD (page file) to create space for the new app. Not only does this extra step slow things (not just an extra step, also this step is in the slowest part of the chain), the Algorithm used is pretty good... but not flawless. Chances are you really needed that second program, so memory gets offloaded again.. etc. etc. etc.
So there are a few very good reasons to get more ram, heavy multitasking, VM's for instance and some heavy duty rendering if the program supports it. That's here you will notice the difference. But in webbrowsing or even HTPC duties? Nothing will change.
More is better... if and when you use it.
*Note: I'm not saying you don't need it, just saying that your initial test method is flawed. Best way to find out the difference is taxing it by opening every app you have^^.