Thoughts on Intel switch mere months before Merom

So what are people thoughts on the fact that Apple goes and switches architecture less than a year before yet another architecture change (PowerPC to x86 32 to x86 64)? I was considering picking up a MBP as my first Mac, but it concerns me that the binaries produced over the next few years will leave the small time period of x86 32-bit out in the cold, since the majorities of Macs will be PowerPC and x86 64 (Merom is planned for ~August I believe).

Just wondering peoples' thoughts on this!

 

Tiger already is a 64-bit OS, so it'll give those who have 64-bit CPUs a performance boost. In addition the PPC G5 was a 64-bit CPU, so those who have that and Tiger are already running 64-bit OSes.

I don't think that 32-bit will die off anytime soon, (even in the Mac world, which usually moves faster than the Windows world) as you stated, the majoriy of users will still be using 32-bit technology when Intel releases Merom. I estimate it'll take a good 2-3 years after the release of Merom for 64-bit to be mainstream.

 

Most applications will not really benefit from being written in 64-bit code compared to 32-bit. The ones that do, you shouldn't be running on a laptop anyway.

Science applications and other number crunching programs will benefit greatly, however the real question is, how many of those do you use? You will get a much higher performance gain when programs become multi-threaded and can take advantage of both processor cores.

Conroe/Merom/Woodcrest are brilliant chips and offers the best performance you can buy at the time of release.

 

the 64 bit revolution has been "coming" for 2-3 years now

i doubt we will see the massive movement to 64 bit programming for at least another 3-5 years

and by then we will all be buying new laptops

 

64-bit isn't going to be anything great. There will be NO performance advantages for the average user. The only, veeeery minor thing the average consumer may like is the fact that you can run 4gb+ of ram.

Just to make it clear, 64-bit just means the cpu can address longer lines of code. When a 32-bit computer gets a piece of code longer than it can handle, it chops it up into bits. What could be one process on a 64-bit cpu may take two processes on a 32-bit cpu.

The million dollar question is: "When does the cpu really see sections of code over 32-bits?"

I'm no computer expert, but I'd assume rarely...unless you're running mathmatica or some industry-grade software. 64-bit CPU's, as of now, are a marketing gimmick to the average consumer. Bigger doesn't always mean better (some 64-bit versions of 32-bit programs actually run slower).

Another thing, updates in the computer world always happen at this rate. The trick is to not get envious of new product releases. Buy what you want and enjoy it for at least a couple of years. There will always be something newer and nicer...

Apple is just staying competitive now that they are on the same playing level as the average PC. No longer can they hide behind the fact that Apples and PC's are incomparable...because with Intel, they pretty much are (with regards to speed).

 

You know whats the problem? The problem (i think) is that anyone who hears of 64 bit, would think on a sudden 2x performance increase. But thats not true.

MacOSX is not a full 64 bit OS. In fact, there are G4 processors (the 603 series) which had an ALU of 32 bit, and if needed, would take 2 cpu cycles to accomplish a 64 bit instruction.

Indeed, 64 bit is the way to go, but it wont bump performance miracolously, if you got a 32 bit int and a 64 bit int, with the value of "1-4 thousand million" (which is a 32 bit integrer) what would take more to the processor's bus to transfer? 32 bits or 64 bits?

A 64 bit integrer would raise accuracy from 4,294,967,296 to 18,446,744,073,709,551,616 (try to read that number, a hell lot in a few words lol) thats counting non signed ints.

As LostCause says, some 64-bit versions of 32-bit programs actually run slower, and a quote from wikipedia:

....Theoretically, some programs could well be slower in 64-bit mode. Under some architectures, instructions for 64-bit computing take up more storage space than the earlier 32-bit ones, so it is possible that some 32-bit programs will fit into the CPU's high-speed cache while equivalent 64-bit programs will not. In basic terms moving 64 bits at a time to perform otherwise 32 bit work simply requires more processing effort to/from memory. A common argument is that, in applications like scientific computing, the data being processed often fits naturally in 64-bit chunks corresponding to double-precision floating-point types, and will be faster on a 64-bit architecture because the CPU will be designed to process such information directly rather than requiring the program to perform multiple steps — this is erroneous, however, because most 32-bit CPUs already have a 64-bit wide data bus and 64-bit registers for floating-point quantities. The only speed advantages come for manipulating 64-bit integer quantities, but this is rarely a performance-limiting task even for applications (such as large-file I/O) that require such manipulations......

http://en.wikipedia.org/wiki/64_bit