• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
Again this kind of ignore the entire 'there is no magic flag that makes a working 64-bit exe pop out the other side'
....no?

I never said such a thing. People argued the benefits of 64bit as if it's not a sure thing. Supporting 64bit has to happen eventually like it or not, it's pretty stupid to argue otherwise to be honest

Even if you're trying to be stubborn just because, assuming your program uses UTC time system it will break afted January 19 2038, you'll have to support 64bit by then
 
64-bit intel has access to a lot more CPU registers than 32-bit intel, which compilers can make good use of to optimise the resulting machine code. So even if you're not using any numbers with better than 32-bit precision, or need more than 4Gig of RAM, you should still get a noticeable (5% - 10% IIRC depending on exact workload) speed boost migrating to x64. Apple dropping 32-bit also makes their own lives easier, and 64-bit only Macs should be faster & more power efficient.

There's a bunch of good reasons for Apple to drop 32-bit support. Dropping OpenGL & ignoring Vulcan, that's where I'm finding it a lot harder to support their reasoning.... ;-)
 
Also,migrate to X64 double the pointers size,so it does not necessarily increase the performance.
That's actually a moot point. When you run 32 bit on a 64 CPU architecture, the CPU's microcode will move everything around as 64 bits, regardless if you use only 32 bits from the data or not. Likewise, all the cache lines of the L1, L2 and L3 caches will be 64 bit based. And memory transfers to and from the CPU will be in 64 bits chunks, as the data bus to the memory DIMMS is 64 bits wide.

In other words, on a 64 bits CPU, all data transfers to and from the CPU will always be performed as 64 bits bus transactions, regardless if you use all 64 bits of the value, or less. Using 32 values for pointers doesn't make it faster, nor does using 64 bit points make it slower.
 
That's actually a moot point. When you run 32 bit on a 64 CPU architecture, the CPU's microcode will move everything around as 64 bits, regardless if you use only 32 bits from the data or not. Likewise, all the cache lines of the L1, L2 and L3 caches will be 64 bit based. And memory transfers to and from the CPU will be in 64 bits chunks, as the data bus to the memory DIMMS is 64 bits wide.

In other words, on a 64 bits CPU, all data transfers to and from the CPU will always be performed as 64 bits bus transactions, regardless if you use all 64 bits of the value, or less. Using 32 values for pointers doesn't make it faster, nor does using 64 bit points make it slower.
Ok thanks for your reply. I am not a programmer,so with your explanation,i understand that i say mistakes.It allows me to learn more about programming.Sorry for my mistakes.
Cordially.