[codeflo]

We already have 128-bit registers for vector computations (for example, the SSE registers in x86).

As for addressing 128 bits of memory: that's more than a century off even if memory continues to double every two years (which doesn't seem likely to begin with). It's actually plausible that the step to 64-bit addressing was the last one, ever.

[colanderman]

Don't know why you were downvoted. Modern x86 chips not only have 128-bit vector registers, but also 256-bit ones: http://en.wikipedia.org/wiki/Advanced_Vector_Extensions

(And while one could argue that a vector register isn't a "real" general-purpose register, the primary difference is that you can't perform fullwidth arithmetic, which is of very little use > 64 bits anyway.)

[dragontamer]

64-bit ARM means passing around 64-bit pointers by default.

In most code settings, this pollutes the cache (you can only hold half as many pointers), and leads to slightly weaker performance.

64-bit computing is NOT an advantage in the phone world. It is a massive advantage for database applications or large web services... but certainly not for phone apps in the near future.

[codeflo]

I know what 64-bit computing is. And I didn't say anything about phone chips, or ARM. It seems like you're posting this rant to every subthread, regardless of content?

[data-cat]

That is an interesting conclusion. I did not realize x86 already had 128 bit registers. That was really what I was referring to. We might not need to worry about addressing more memory yet but I feel like larger registers will still be useful.

[stephencanon]

Actually, x86 already has 256-bit vector registers (AVX), and will have 512-bit vector registers in a few years (AVX-512). If you include the exotic-but-still-x86-based xeon phi it already has 512-bit vector registers.