Reply To: Splain me 32 bit vs 64 bit
Wikipedia mentions that the advantage of 64 bit systems is mostly seen when working with large data sets in applications such as digital video, scientific computing, and large databases.
I’d like to point out that 32 bit systems are not really limited to 4GB of RAM. There is a feature called physical address extension which allows them to use up to 64GB of RAM, albeit with some overhead. I’m still using a 32 bit operating system and I’ve got 6GB of RAM. If I ever get around to re-installing, I’ll switch to 64 bit.
If every location and every hunk of data is now going to be 64 bits wide, doesn’t that effectively cut your memory in half?
Not necessarily. If you have a large chunk of data to store, the 64 bit architecture will simply use half as many 64 bit locations to store it. If the data set doesn’t divide evenly into 64 bits though, the wasted padding will obviously be larger than with a 32 bit system, but I don’t think this is a big deal in practice. Also, I think the 64 bit term applies only to the processor’s register size. I think RAM is usually addressed as individual bits, not as 64 bit chunks.
Disk space is addressed in much larger chunks (see filesystem block size). 4096 bytes is typical, so there’s no waste with a 64 bit system, and it will much faster reading the data since it needs half the number of clock cycles to do so.
I don’t think the programs themselves will be take up more space. That was an issue in the RISC vs CISC debate, but Intel’s dominance has rendered it moot.