flat assembler
Message board for the users of flat assembler.
Index
> Main > [] |
Author |
|
Spool 12 Jan 2013, 16:55
[ Post removed by author. ]
Last edited by Spool on 17 Mar 2013, 04:07; edited 2 times in total |
|||
12 Jan 2013, 16:55 |
|
HaHaAnonymous 12 Jan 2013, 17:11
[ Post removed by author. ]
Last edited by HaHaAnonymous on 28 Feb 2015, 22:02; edited 2 times in total |
|||
12 Jan 2013, 17:11 |
|
Spool 12 Jan 2013, 17:18
[ Post removed by author. ]
Last edited by Spool on 17 Mar 2013, 03:51; edited 1 time in total |
|||
12 Jan 2013, 17:18 |
|
HaHaAnonymous 12 Jan 2013, 17:35
[ Post removed by author. ]
Last edited by HaHaAnonymous on 28 Feb 2015, 22:02; edited 1 time in total |
|||
12 Jan 2013, 17:35 |
|
JohnFound 12 Jan 2013, 20:18
Bargest, "inc [memory]" takes 2 memory operations as well. Because the CPU can not do arithmetic directly in the memory. The value have to be read, processed and then write. Of course the cash memory can affect this process as well. (although I also think "inc [memory] faster" because at least only one instruction have to be read from the memory, decoded and executed. Well, from the cash actually.
|
|||
12 Jan 2013, 20:18 |
|
Bargest 12 Jan 2013, 20:24
Quote: Bargest, "inc [memory]" takes 2 memory operations as well. I know this. But: 1) It is one operation, so just one opcode is peeked out from buffer and decoded; 2) I think in this instruction address is set on address bus just for one time (it is one of the longest parts of memory operations). So I consider it is 1.5 memory operation. And cache improves speed of first case as well. |
|||
12 Jan 2013, 20:24 |
|
< Last Thread | Next Thread > |
Forum Rules:
|
Copyright © 1999-2024, Tomasz Grysztar. Also on GitHub, YouTube.
Website powered by rwasa.