Home Forums Hardware Mac Desktops & Notebooks RAM upgrade question

I just bought a 4Gb DIMM to upgrade my deuced Mac(hina) from 4 to 6GB, and the move will leave me with a spare 2Gb DIMM.

My daughter's got a Late 2008 MacBook that takes the same DIMMS as mine, and I'm thinking about giving her the spare to upgrade her RAM from 2 to 3Gb if it's viable.

Am I making a mistake by relying on this?


Thanks.

FWIW Small Dog's observations match mine. Not even a "power user" can accurately perceive a speed difference between matched and unmatched RAM without running sensitive speed tests. I think the speed differential used to be somewhat more noticeable when memory speeds and processor speeds were a LOT slower than today's. That said, given a choice I will usually buy matched DIMMs, but that is probably more out of habit as any thing else.

Thanks.

Matched DIMMs would be my choice, too, but that would mean buying a pair, as opposed to just swapping out a 1Gb for a free 2GB.

A mismatched 3Gb set doesn't seem like it'd be any more of an issue than the mismatched 6Gb set that MacManiac has had such success with, although it doesn't sound as though it will be as successful an upgrade.

Why would you think that? If it's because the amount of RAM added is smaller than in MM's case, consider that your daughter's MB has a smaller amount of RAM to begin with, and arguably stands to gain more* by an expansion despite the overall limitation. But the bottom line is that you won't know until you try.

*) Obviously, the precise 'numerical' improvement, however measured, remains to be seen, but that's not what I mean here.

Well, both MacManiac and I have upgraded similar MacBooks from 2GB of RAM to 4 without seeing any differences particularly worth crowing about, so it seems likely that a 2 to 3Gb upgrade, while it will certainly be helpful, won't be any more notable.

Do you recall ever having seen as exuberant a report about a RAM upgrade as MacManiac's?

I've seen similar RAM expansion recommendations (they're fairly standard), if not the same exuberant improvement reports (they're usually implied). But the size of the improvement must depend to a large degree on the task(s) affected, so it might help to ask MM exactly which one(s) went that much better in his case, and how much better.
Likewise, the possibility of not having run a critically RAM-dependent task on previously upgraded MBs may have influenced both your judgment there (just thinking out 'loud').

In my case the most immediately noticeable improvement came with the opening of Safari....it opens significantly faster and loads pages with notably less waiting.

The Microsoft Office 2008 suite is also noticeably more responsive.

I'm not sure why this is, but I am certainly impressed by the change in at least this immediate area. I haven't run any formal benchmarks to evaluate empirical measurements, so please take my enthusiasm for what it's worth.

Well, I installed my new DIMM a while ago, and I suppose its a function of the apps and processes I run, but I haven't noticed any "Wow!" improvement.

Safari does launch and load pages more quickly, but the most noticeable change I've noticed so far is that the "Free" segment of my MemoryStick bar is considerably larger. (I'm sure I'll see much improvement when I get to editing some audio files.)

XBench didn't show any improvement at all, but I'm not certain that it measures anything pertinent.

Edit: I forgot to mention that I tried the DIMM in both slots and didn't notice any particular difference.

.....I guess that's why the caveat "YMMV" gets used so much.

You may need to clear caches using the SafeBoot (holding the shift key during start) and then reset PRAM to clear any legacy instructions before the additional RAM can provide the same benefit that seems to happen for me.

Are you in Lion?

I concur that XBench does not measure anything pertinent. It tests hardware speed, pure and simple and while that is definitely a major component of system performance it does not test anything that would be effected one way or the other by additional RAM. (FWIW all the benchmark apps listed on MacUpdate or the App Store are pure hardware benchmarks.) The improvement expected with additional RAM occurs when RAM gets filled up and the system is forced to page to the HD. More RAM makes it less likely your system will be forced to page. The WOW factor occurs when you have been running lots of memory intensive apps at the same time and your system has been paging often (constantly?) because of limited RAM.

My wife has yet to meet the app that she is willing to close so she often has seven or eight or more major apps running at the same time. Adding more RAM to her MacMini made a huge difference in her perceived performance. Truthfully I am not sure her actual throughput increased that much, but there are a lot fewer annoying pauses and inexplicable slowdowns while she is working.

Of course all of this has been helped by improvements Apple has made in virtual memory management with almost every upgrade of OS X which not only reduces the need for paging but makes paging less noticeable to the user. At the same time Apple has increased the standard and recommended minimum RAM several times as well.

> My daughter's got a Late 2008 MacBook that takes the same DIMMS as mine, and I'm thinking about giving her the spare to upgrade her RAM from 2 to 3Gb if it's viable.

Well, it turns out that I got her to "max" (4Gb) out her RAM when she bought her MacBook, so my initial question has become inconsequential specifically, although not generally, and I've got no test results to post.

Thanks for the suggestions.

I both Safe Booted and reset PRAM with no resultant significant changes, so I guess my unspectacular upgrade results are just a function of the non-RAM-intensive apps I run virtually 100% of the time.

I note, though, that editing music files in Fission has not yet generated any swap files, as it usually does.

Nope, I'm not in Lion...still 10.6.8. (I thought our Macbooks could run 64 bit apps but couldn't boot into 64 bit mode under any circumstances, or was that only as respects Snow Leopard?)

Thanks for confirming my suspicions.

I pretty much realized that XBench measures only hardware speed, but I wasn't certain whether or not it makes intensive use of RAM in the process.

Just to see what actually happens when XBench is running I ran it with Activity Monitor open, and I found that although its "Real" memory usage is not at all out of line...it never even gets to the top of the list, its virtual memory (which I realize has nothing to do with RAM) usage, which approaches 2Gb during this part of the test, ASTONISHINGLY hits 16Eb during this part. (It's a wonder that my HD didn't explode into a mushroom cloud of radioactive data! )

Well, I just finished a major music editing project, and I guess my extra RAM has finally showed its muscle.

By the end of the project, which was completed with no resultant swap files (but with some annoying, although not hair-tearing, beach-balling), MemoryStick showed less than 800Mb of free RAM, which suggests to me that my last 2Gb was put to good use.

Even if that's the sum total of the benefit to be derived, I suspect that the DIMM was worth the price.