Return Styles: Pseud0ch, Terminal, Valhalla, NES, Geocities, Blue Moon. Entire thread

optimizing CPU/FPU instructions

Name: Anonymous 2013-10-07 16:13

is it possible to code, say, an ADD that's faster than the actual instruction for ADD, only using other CPU functions (no physical access to the chip)? or are all CPU instructions always perfectly optimized?

Name: Anonymous 2013-10-07 16:32

All CPU instructions are made of NOR gates.  And yes, they are perfectly optimized. However, you can sometimes use LEA in place of ADD to get a ``free'' addition executed in parallel to other code.  See some explanation here http://stackoverflow.com/a/6328441/1116279 and also check the comments.

Name: Anonymous 2013-10-07 16:35

>>2
All CPU instructions are made of NOR gates.
Really? In every architecture? Is using only NOR the optimal choice, or is it just a matter of convenience using the same gate over and over, knowing it's Turing-complete? Is NOR better than NAND?

Name: Anonymous 2013-10-07 20:03

c

Name: Anonymous 2013-10-07 20:03

h

Name: Anonymous 2013-10-07 20:04

e

Name: Anonymous 2013-10-07 20:04

c

Name: Anonymous 2013-10-07 20:04

k

Name: Anonymous 2013-10-07 20:04

'

Name: Anonymous 2013-10-07 20:05

e

Name: Anonymous 2013-10-07 20:05

m

Name: Anonymous 2013-10-07 21:17

>>4
>>5
>>6
>>7
>>8
>>9
>>10
>>11
>le pedophile sage

Name: Cudder !MhMRSATORI!fR8duoqGZdD/iE5 2013-10-09 7:33

is it possible to code, say, an ADD that's faster than the actual instruction for ADD
No. If it is then your CPU needs redesign.

Name: Anonymous 2013-10-10 9:51

>>13
fsincos+fdiv is much faster than fptan
pwned

Name: Anonymous 2013-10-10 10:06

>>13
>le pedophile sage

Name: Cudder !MhMRSATORI!fR8duoqGZdD/iE5 2013-10-10 10:29

>>14
Not that much faster (in my experience, the low values for FPU instructions are very rare in practice, and only occur on "round" numbers and special cases like 0/1.)

Nehalem:
FSINCOS ~110
FDIV 7-27
FPTAN ~115

Sandy Bridge:
FSINCOS 43-123
FDIV 10-24
FPTAN ~130

Ivy Bridge:
FDIV 10-24
FSINCOS 50-123
FPTAN ~130

Not the same precision either due to rounding and such.

Name: Anonymous 2013-10-10 11:04

>>16
>le pedophile sage

Name: LE NEGATIVE INFORMATION 2013-10-10 14:04

>>16
the point is that fsincos can be fast but fptan is always slow

#yoloswag
#haxmyanus
#progsnake
#spudrospärde:-----DDDD
#thenailthatsticksoutthemostgetshammereddownthefastest
#wantonpseudointellectuals
#iwannaCAOS!

Name: Anonymous 2013-10-10 19:40

>>18
>le pedophile email feild contents

Name: Anonymous 2013-12-22 9:52

YOU HAVE BEEN VISITED BY LE GREEN SAD NEGRO FROGE OF SADDNESS
REPOST THIS IN 100`000 threads or be a frog!

`
           ████████     ██████       
         █░░░░░░░░██ ██░░░░░░█      
        █░░░░░░░░░░░█░░░░░░░░░█     
       █░░░░░░░███░░░█░░░░░░░░░█    
       █░░░░███░░░███░█░░░████░█    
      █░░░██░░░░░░░░███░██░░░░██    
     █░░░░░░░░░░░░░░░░░█░░░░░░░░███ 
    █░░░░░░░░░░░░░██████░░░░░████░░█
    █░░░░░░░░░█████░░░████░░██░░██░░█
   ██░░░░░░░███░░░░░░░░░░█░░░░░░░░███
  █░░░░░░░░░░░░░░█████████░░░█████████
 █░░░░░░░░░░█████ ████   ████ █████   █
 █░░░░░░░░░░█     █ ███  █    ███ █   █
█░░░░░░░░░░░░█   ████ ████   ██ ██████
░░░░░░░░░░░░░█████████░░░████████░░░█
░░░░░░░░░░░░░░░░█░░░░░█░░░░░░░░░░░░█
░░░░░░░░░░░░░░░░░░░░██░░░░█░░░░░░██ 
░░░░░░░░░░░░░░░░░░██░░░░░░░███████  
░░░░░░░░░░░░░░░░██░░░░░░░░░░█░░░░░█ 
░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█
░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█
░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█
░░░░░░░░░░░█████████░░░░░░░░░░░░░░██
░░░░░░░░░░█▒▒▒▒▒▒▒▒███████████████▒▒█
░░░░░░░░░█▒▒███████▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒█
░░░░░░░░░█▒▒▒▒▒▒▒▒▒█████████████████
░░░░░░░░░░████████▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒█
░░░░░░░░░░░░░░░░░░██████████████████
░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█    
██░░░░░░░░░░░░░░░░░░░░░░░░░░░██     
▓██░░░░░░░░░░░░░░░░░░░░░░░░██       
▓▓▓███░░░░░░░░░░░░░░░░░░░░█         
▓▓▓▓▓▓███░░░░░░░░░░░░░░░██          
▓▓▓▓▓▓▓▓▓███████████████▓▓█         
▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓██       
▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓█       
▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓█

Newer Posts
Don't change these.
Name: Email:
Entire Thread Thread List