[DBP] increasing a dword so that it doesn't wrap around!

Author

Message

Neuro Fuzzy

17

Years of Service

User Offline

Joined: 11th Jun 2007

Location:

Posted: 19th Oct 2010 11:28

Link

There's a problem with incrementing certain datatypes... After a while, all the bits flip back around. Say you have an eight bit number:
11111110
that number is 254.
11111110+1=11111111 = 255 base ten
however since we only have eight bits to store our number in...
11111111+1=00000000 = 0
in other words, 255+1=0.

To prevent this, you have to take extra precautions; like this code I made for incrementing Dwords, so (with the analogous case with our eight bit number) 255+1=255 and not 0, and 0-1=0, and not 255.

+ Code Snippet

i1 as dword
i2 as dword
d1 as dword
d2 as dword

i1=967295
i2=4294960000
d1=12000
d2=4294967295

do
   if spacekey()
      i1=inc2(i1,i1)
      i2=inc2(i2,1)
      d1=dec2(d1,1)
      d2=dec2(d2,967295)
      cls
      text 0,0,str$(i1)
      text 0,16,str$(i2)
      text 0,32,str$(d1)
      text 0,48,str$(d2)
      sync
   endif
loop



function dec2(a as dword, b as dword) `a-b
   q as dword
   ret as dword
   ret=0
   q=a-b
   if difference(a,q)=b then ret=q
endfunction ret
function inc2(a as dword, b as dword) `a+b
   q as dword
   ret as dword
   ret=4294967295
   q=a+b
   if difference(a,q)=b then ret=q `if q has not rolled over.
endfunction ret
function difference(a as dword, b as dword)
   ret as dword
   if a>b
      ret=a-b
   else
      ret=b-a
   endif
endfunction ret

Useful if you want to...

uh... I'll get back to you on that.

Back to top

Profile PM Email

HowDo

21

Years of Service

User Offline

Joined: 28th Nov 2002

Location: United Kingdom

Posted: 20th Oct 2010 18:30

Link

good one this, but it does not work with negative numbers, could it?

Dark Physics makes any hot drink go cold.

Back to top

Profile PM

Benjamin

21

Years of Service

User Offline

Joined: 24th Nov 2002

Location: France

Posted: 20th Oct 2010 22:08

Link

Quote: "Useful if you want to...

uh... I'll get back to you on that."

Back to top

Profile PM Email

Neuro Fuzzy

17

Years of Service

User Offline

Joined: 11th Jun 2007

Location:

Posted: 20th Oct 2010 23:17 Edited at: 20th Oct 2010 23:17

Link

Actually, I think it can work with integer too!
i can't test it out right now, but this should work:

+ Code Snippet

function dec2(a as integer, b as integer) `a-b
   q as integer
   ret as integer
   ret=−2147483648
   q=a-b
   if difference(a,q)=b then ret=q
endfunction ret
function inc2(a as integer, b as integer) `a+b
   q as integer
   ret as integer
   ret=2147483647
   q=a+b
   if difference(a,q)=b then ret=q `if q has not rolled over.
endfunction ret
function difference(a as integer, b as integer)
   ret as integer
   if a>b
      ret=a-b
   else
      ret=b-a
   endif
endfunction ret

Actually, whenever I write some quick and dirty code to convert variables to RGB values, I use bytes and cap them at 255 using similar code.

for any signed or unsigned binary number, this should work (reeplacing the placeholders):

+ Code Snippet

function dec2(a as datatype, b as datatype) `a-b
   q as datatype
   ret as datatype
   ret=datatype_minimumvalue
   q=a-b
   if difference(a,q)=b then ret=q
endfunction ret
function inc2(a as datatype, b as datatype) `a+b
   q as datatype
   ret as datatype
   ret=datatype_maximumvalue
   q=a+b
   if difference(a,q)=b then ret=q `if q has not rolled over.
endfunction ret
function difference(a as datatype, b as datatype)
   ret as datatype
   if a>b
      ret=a-b
   else
      ret=b-a
   endif
endfunction ret

probably won't work with floats tho.

Back to top

Profile PM Email

Sorry your browser is not supported!

Code Snippets / [DBP] increasing a dword so that it doesn't wrap around!