Category talk:ND-100 instructions: Difference between revisions

From NDWiki
Jump to navigation Jump to search
(About the 'Register notation uniformity fix' changes)
 
(Great)
 
(One intermediate revision by one other user not shown)
Line 6: Line 6:
  (ea) means the value of the memory location pointed to by ea, not ea itself
  (ea) means the value of the memory location pointed to by ea, not ea itself
'ea' was used consistently that way already, now it's done for registers too. 'ea' is like a pseudo-register in this respect (in reality it is a physical register of course - the CPU sets it up before accessing memory)
'ea' was used consistently that way already, now it's done for registers too. 'ea' is like a pseudo-register in this respect (in reality it is a physical register of course - the CPU sets it up before accessing memory)
Other things that could be fixed (noticed along the way):
Other things that could be fixed (noticed along the way):
* Use of = vs := (we should e.g. settle on the latter)
* Use of = vs := (we should e.g. settle on the latter)
Line 13: Line 14:
   
   
--[[User:TArntsen|TArntsen]] 10:03, 21 March 2013 (CET)
--[[User:TArntsen|TArntsen]] 10:03, 21 March 2013 (CET)
:Great! I think an additional field is the best way to go, for now. /[[User:Mike|Mike]] 17:05, 22 March 2013 (CET)

Latest revision as of 16:05, 22 March 2013

I went ahead and straightened out the instruction register notation format in the infoboxes. There was a mix of methods, sometimes even inside the same infobox. The following notation is now used for all instructions:

X    means the value of register X
(X)  means the value of the memory location pointed to by X, not X itself
ea   means effective address, the actual address
(ea) means the value of the memory location pointed to by ea, not ea itself

'ea' was used consistently that way already, now it's done for registers too. 'ea' is like a pseudo-register in this respect (in reality it is a physical register of course - the CPU sets it up before accessing memory)

Other things that could be fixed (noticed along the way):

  • Use of = vs := (we should e.g. settle on the latter)
  • Sometimes the 'Affected' infobox field is used to describe the actual function (e.g. 'if X=1 then A=(ea)'), but this is not always feasible, so others just lists registers
  • Sometimes flags affected are listed as well

So maybe we need an additional field which pseudo-explains the operation, and the Affected field could just be a list like it's sometimes used. Or we figure out an even better way.

--TArntsen 10:03, 21 March 2013 (CET)

Great! I think an additional field is the best way to go, for now. /Mike 17:05, 22 March 2013 (CET)