[mrmister]

It's a total indulgence, I'll admit--I just bought this 15" PowerBook 1.5 Ghz slightly used, and it came with 1 gig of RAM, which from experience on my TiBook is enough to be totally servicable--great, in fact. I don't cruch video (often) or do much Photoshop...I mainly just write.

But...why be servicable when one can be extravagant? And since I have a ridiculously overpowered machine already, at least for word processing, why not go whole hog?

After a lot of hemming and hawing I ordered the KingMax RAM that many have been discussing here. Many have posted links to the RAM, but here's another:

http://www.newegg.com/app/viewproduc...111&DEPA=0

The retails at NewEgg for $191 a stick, and shipped for me right away...in fact, I received it one day after ordering in the early morning.

The AlBook has Apple-installed RAM, branded SAMSUNG, and appears to be almost identical to the KingMax RAM in size and thickness--these aren't "stacked" chips like used to happen with older SDRAM.

Installation was very easy. I always found TiBook RAM installation easy as well, but I love that the RAM slot on the bottom allows the laptop to have a firmer keyboard. It seemed like it might be harder to remove the RAM due to the small size of the install space, but it turned out to be simple.

Once installed I rebooted, and Apple System Profiler recognized the RAM right away. I ran the extended hardware test twice, and TechTool 4 as well--all are reporting the RAM as good, and so far I've seen no anomalous behavior.
Temperture Monitor is reporting the system is running at the same temperture as before--I see variations up and down a degree, but nothing to write home about.

I'm also running the excellent MemTest, a wonderful Linux app that thoroughly thrashes new RAM to ensure it's working.

http://www.macupdate.com/info.php/id/14004

(It's Linuxy, so you need to follow the instructions, but it's worth it.)

In real world terms the results are more noticable than I expected. Apps are faster for switching, the GUI is more responsive--I didn't really expect to see much of a change. The coolest part is how there seems to be no change in system responsiveness as I ladle apps up--I have 23 running now, ripping a CD, rebuilding my Entourage database and typing this message, and there's no change. I always end up killing off my apps out of habit, but I may leave more of them running now that I can't seem to use up my RAM.

Economically, it cost me $400, but I'm reselling my 2 512mb chips for $100 apiece once I am done testing, so it works out to be a $200 upgrade for me. Well worth it so far.

[jamil5454]

That Rawks.

[mrmister]

There's also a GUI version of memtest, for those who fear and loathe the command line:

http://www.macupdate.com/info.php/id/15837

Doesn't test as much memory, because you obviously can't run this GUI app in single-user mode, but still useful.

[mrmister]

Just let the PowerBook crunch on memtest in single-user mode for 12 hours--no glitches or problems.

[mrmister]

Inbetween work today I cleared my PRAM and did battery tests of the stock Apple RAM (1 gig) vs the KingMax (2 gigs). I expected a hit in battery life due to there being more RAM, but had hoped it wasn't too extreme.

The results? No difference. After two tests, 1 gig of Samsung yields a runtime (screen at full, energy saver settings on reduced with everything conservative, Airport on, 6 or so apps running internet, mail, IM) of 2 hrs 35 mins. The 2 gigs under the same conditions gave me 2 hrs 43 mins...which seems odd, that it would be more, but it's close enough with enough variables that I think they're basically the same.

I mean,don't get me wrong: they both SUCK for battery life, but the RAM doesn't seem to change anything.

I do know from experience going from 512 to 1 gig of RAM back in 2001 with a Titanium 500 that more RAM means more battery degrading during sleep, because electricity needs to be trickled into more memory. I haven't had an opportunity to test that yet.

Why am I being so thorough? Because every day 6 people ask about 1 gig memory modules, so someone should see if they can figure this voodoo science out. For example, I am very surprised to find no difference in run times between Samsung "low power/high cost" RAM and the RAM I bought from NewEgg.