RAM latency for laptops makes no difference. Only if you run synthetic benchmarks are you going to notice the difference. However, in real world, no difference. However, don't buy CL 10 RAM, that will be noticeable.
Oh, ok. I was reading some stuff online about how people overclock their RAMs to make things "snappier" and all, but I don't want to mess with that, so I was wondering how come the old DDR2 RAMs for Macbooks had a low latency option but none for DDR3 RAMs.
On a RAM related topic, there was a thread I caught here in the forums a few days back about how MAC OS X starts with the 32 bit kernel and you have to hold down some buttons for it to start with the 64 bit kernel, and people were talking about RAM limitations between the kernels. Can the 32 bit kernel use the full 8GBs of RAM that we can upgrade to? That part was a bit confusing for me..
The current kernel for OS X is 32-bit. Meaning the kernel will never access more than 3.25GB of RAM. That is no problem since the kernel barely goes over 20MB RAM footprint. However, the rest of the system, applications and processes do need space. Because of that OS X was (or rather is being) transitioned into 64-bit. Almost all (if not all) current Kernel extensions (kexts) are 64-bit. Furthermore, many native OS X applications are already 64-bit. That means they can use more than 3.25GB RAM without problems. Also, any application that needs 64-bit support will find it due to the revised kexts.
Ok, let's see if I get this right - the current 32 bit kernel with today's apps, many of whom are 64 bit, can in fact make use of the full 8GBs of RAM? Is that right?
I think the lowest I've see on the iStat pro widget was ~1.7GB available (rest was used or tied up). It usually floats around 2GB-2.5GB free/available. Has anyone with the 8GB upgrade observed the extra RAM being used with the 32 bit kernel running the apps?