I got a new stock 12" iBook about a month ago. I hobbled along for that period of time using the standard 256MB of RAM, intending to upgrade that as soon as possible. When I got my September paycheck, I immediately ordered two 512MB RAM modules (one for the iBook and one for our new 12" Powerbook). Since I've read a few threads on here that said the PC2700 RAM is fine for iBooks, I went ahead and purchased two identical PC2700 chips from macsolutions.com. I should mention that the description of the PC2700 RAM on their website also stated it was suitable for iBook G4s. Anyhow, I got the RAM express delivered and popped one of them in my iBook. I ran a pre- and post-upgrade Xbench, and went from an overall score of 78.59 to 88.52. Okay, fine. But then I looked a little closer and realized that the total RAM being reported was only 512, rather than the 768 (256 base + 512 upgrade) it should have been. Weird. I took the RAM out and re-inserted it to make sure it was seated properly, but it was still only being recognized as a 256MB chip. Rats. I began to wonder if perhaps PC2700 really couldn't be used in iBooks. After several tries of removing and re-inserting the memory, I ended up trying the second RAM module that had been destined for the Powerbook. Popped it in, took a look at the system profile, and was happy to see that it was properly reporting 768MB of RAM. Woohoo! I took the first RAM module (the one the iBook kept insisting was 256MB) and stuck it in the Powerbook. It is also recognized as a 512MB chip. Sweet. Now here's where the weirdness sets in a bit. I went back to the iBook and decided to do an Xbench report on it, now that the correct amount of memory was being recognized. But for some odd reason, I actually scored lower than I had when the system was only recognizing the chip as 256MB! The overall difference between the two was 88.52 (when recognized as 256MB) down to 80.13 (when recognized as 512MB). Hmmph! I decided I was in an adventurous mood, and swapped the modules one final time. Again, the iBook performed better (according to XBench) with the chip it recognized as 256MB. Curiouser and curiouser. Now, of course, I realize that Xbench isn't infallible. But I just think it's odd that it's giving me repeatable results that say the module properly recognized as 512MB is barely increasing my scores from the stock memory the iBook shipped with. Yet, it performed better (in some areas of the test, TWICE as good) with the RAM of which it was only recognizing half. What gives?? I realize that OS X is "picky about RAM" because I've read that about a zillion times on here. I'm just confused why, in this case, it seems to like RAM that it mis-reads the best!