Mailinglist Archive: opensuse (686 mails)

< Previous Next >
Re: [opensuse] 64 bit vs 32 bit RAM consumption
Carlos E. R. said the following on 07/23/2013 07:01 PM:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256



El 2013-07-23 a las 15:44 -0400, Anton Aylward escribió:

Werner Flamme said the following on 07/23/2013 02:45 PM:


If you have to add 2 and 3 it will take the same time on both archs. 64
bit is not faster, it is larger. Compare a road truck of 6 wheels to
another of 12 wheels and double length.

Huh? That doesn't make sense to me?
Can you explain it it detail - and drop the analogy.

Give op-code example. Doesn't have to be from a real machine, just
hypothetical.


No, I will not bother. If you don't understand it, I will not bother to
explain.

I think we're out of sequence here, Carlos. I'm in agreement with you, it was Werner's post I was questioning.

It doesn't matter if the "2" and the "3" are in registers or memory, if its a register-to-register, register-to-memory or memory-to-memory operation. These are small integers. The issue is the op-code set. I couldn't figure what Werner's bit about 'zeros' had to do with it.
The only reason I can' assert all this for an 8-bit machine is that it probably won't have op-codes for all those classes of operations. But yes, 64, 32 and 16 bit machines certainly. (I get to pick and choose, mind.)


Actually, IIR the old Zilog Z-8000 architecture (which AMD produced to Avionic grade and to MIL-SPEC in small quantities) was a 16-bit machine where you could double up the registers and treat the top 8 16-bit registers as 4 32-bit registers and do 32-bit native operations on them.
I suspect, but don't know for sure because I never used it, that the later 32-bit machine, the Z-80,000, could do a similar thing and run in 64-bit mode. Of course this is only arithmetic and logic, not address space.




--
"Context is everything"
--
To unsubscribe, e-mail: opensuse+unsubscribe@xxxxxxxxxxxx
To contact the owner, e-mail: opensuse+owner@xxxxxxxxxxxx

< Previous Next >