|
|
|
| Authored by: BitOBear on Sunday, October 14 2012 @ 10:39 AM EDT |
Most computers are _incpable_ of loading a single bit (e.g. exactly one zero or
one) from any memory source anywhere. And when you _do_ that single bit will be
zero-padded or masked-off when it is read or written (respectively) to/from the
register where it lived with its siblings.
The get single bit instructions are actually somewhere around ten years old and
there are only a couple of them.
Most computers must load at least a byte (eight bits) at once.
Any computer running Windows CE can only load a 32bit word at once (each
character in a string is 32bits long in windows CE and there is no legal
"char" type) This was to "simplify" the instruction set, but
it explains why the platform was such a memory pig. Plus it had to explode
everything it got from the internet from 8bit to 32bit representation by doing
this ugly copy-shift-mask operation on every freaking byte. Go Microsoft Design
Guys! 8-)
So really, you can not operate on bits, only bytes (e.g. symbols) in (almost
all) computers.[ Reply to This | Parent | # ]
|
| |
| Authored by: BitOBear on Sunday, October 14 2012 @ 10:47 AM EDT |
There are _lots_ of instructions that manipulate zero symbols. You have
previously said "they don't count" because they are "just
housekeeping". They do make the difference between programs that work and
thouse that don't.
Those instructions include, but probably aren't limited too:
Memory barriers (read barrier, write barrier, full barrier)
Hlt (non-deterministic "pause", on the intel platform).
Lock (bus locking, e.g. prevent others reading and writing for one instruction
cycle).
NoOp (do nothing for one instruction cycle).
That is, these instructions do not read, write, nor manipulate any symbol at
all, and yet they are necessary to prevent or enable a myriad of behaviors
within real modern computers.
You don't agree. I am not going to bang my head against your wall of denial
again here. I'm not even going to read the response since we have reached
impasse over this before.
Their existence in the real world remains as mute accusation of the non-math
basis of _some_ computer programs whether you agree or not.[ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Sunday, October 14 2012 @ 11:34 AM EDT |
WRONG.
that math is an interreptation of lower math that tells the
device to send certain beams of light and that device is
math hard coded that tells the atoms and molecules in proper mathematical
sequences to do stuff.
one could truly argue that all devices are in fact math
because they are doing a sequential system of math. Go on
USA try and keep your retarded systems cause the rest of the
world
has stopped laughing its not beyond funny its sad....
no wonder your world rank in math is 32...[ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Monday, October 15 2012 @ 03:17 AM EDT |
Computer instructions are of the following types:
1) Do a computation: take the contents of memory cell A, and the contents of
memory cell B, and use them to compute a value which is then put in memory cell
C
2) Wait.
There are a couple of things the computer does *without* instructions.
Input is one of them. Input, in a typical modern computer, just gets dumped
into particular memory cells by the input subsystem. (To see whether they've
changed, software must issue instructions to copy the value of those cells to
other cells on a regular basis.)
Output is another. The output subsystem in a modern computer typically simply
reads a particular set of memory cells and blasts them to the output device at
some interval.
(There are some other I/O methods, but they actually don't add anything
interesting to the process of writing software. It still is just a matter of
writing a completely abstract list of computational steps, and encoding those
steps as numbers for the "Universal Computing Machine" to deal with.)[ Reply to This | Parent | # ]
|
- nope. - Authored by: jesse on Monday, October 15 2012 @ 08:31 AM EDT
- nope. - Authored by: PolR on Monday, October 15 2012 @ 08:58 AM EDT
|
|
|
|
