S
swaroophr
Which of switch statement and if-else statement takes less time to
execute?
execute?
Which of switch statement and if-else statement takes less time to execute?
E
Emmanuel Delahaye
There is no definitve answer to this question. Dd some tests and
measurements in known conditions.
--
Emmanuel
The C-FAQ: http://www.eskimo.com/~scs/C-faq/faq.html
The C-library: http://www.dinkumware.com/refxc.html
"C is a sharp tool"
E
Emmanuel Delahaye
(supersedes <[email protected]>)
There is no definitve answer to this question. Do some tests and
measurements in known conditions.
--
Emmanuel
The C-FAQ: http://www.eskimo.com/~scs/C-faq/faq.html
The C-library: http://www.dinkumware.com/refxc.html
..sig under repair
There is no definitve answer to this question. Do some tests and
measurements in known conditions.
--
Emmanuel
The C-FAQ: http://www.eskimo.com/~scs/C-faq/faq.html
The C-library: http://www.dinkumware.com/refxc.html
..sig under repair
B
Bhan
I am trying the same.Please help me out
Thanks for your reply
Bhanu
Thanks for your reply
Bhanu
A
Anonymous 7843
The difference between them pales in comparison to other
issues such as algorithm choice.
That being said, switch and if-else do different things.
If you have a two-way condition, use if-else.
If you have three or more cases, use switch.
Trying to express a switch as a chain of if-else's is unlikely to be
better than a switch statement. The only time this should be
considered is if you know that one case out of many is statistically
far more likely to occur. Then you might consider testing for that one
case and using switch on the rest.
K
Keith Thompson
Bhan said:
You are trying the same as what? Provide some context. Don't assume
that we can see the article to which you're replying.
And follow the newsgroup a while before posting. If you had done so,
you would have seen the following hundreds of times:
If you want to post a followup via groups.google.com, don't use
the broken "Reply" link at the bottom of the article. Click on
"show options" at the top of the article, then click on the
"Reply" at the bottom of the article headers.
M
Mark
Keith Thompson said:
He didn't have enough time to provide context...
There was only 5 minutes left for him to complete his online quiz
and he needed the answers quick!
Besides, the question was asked in the message subject... can't you read?
C
Christian Bau
Use a profiler. Try both ways and measure the difference in speed.
If you don't have a profiler, call the function you try to improve
hundred times instead of once only and use a stopwatch. Or call it
thousand times. Then measure.
M
Malcolm
Just depends.
A very short switch will probaly be compiled to exactly the same code as the
corresponding if ... else. A longer switch may use a jump table, if the
constants in the case are small enough. Usually this will be a little bit
faster than if ... else.
N
Neo
When the decisions to be made are huge
like 1000x transactions you prefer switch which is efficient and fast
because of its random ness in accessing the content
when the decisions are too less like 10 or so even if the flow goes
through many ifs and else checks its not going to make much a
difference.
so depending on the number of checks
at one point your compiler will take a decision to give out the same
obj code for both .if U manage to use Optimisation.
But giving out an answer like ... which among Switch and If else is
quick
is as meaning less as .. what fills your stomach soon Breakfast or
Lunch ??
It depends on you .and the cook
like 1000x transactions you prefer switch which is efficient and fast
because of its random ness in accessing the content
when the decisions are too less like 10 or so even if the flow goes
through many ifs and else checks its not going to make much a
difference.
so depending on the number of checks
at one point your compiler will take a decision to give out the same
obj code for both .if U manage to use Optimisation.
But giving out an answer like ... which among Switch and If else is
quick
is as meaning less as .. what fills your stomach soon Breakfast or
Lunch ??
It depends on you .and the cook
J
Jack Klein
int main(void)
{
switch (1)
{
}
if (0)
{
}
else
{
}
return 0;
}
Each is twice as faster as the other on a DeathStation 9000.
J
Joe Wright
That's probably the wrong question. The 'switch(a)/case:' thingy is
limited to multi-way decisions based on the integral value of a. The
'if/else if/else' thingy is much more flexible in that the if conditions
are expressions yielding true or false. I don't use switch() much.
P
Peter Pichler
Anonymous said:
....or if the condition may change between tests and you want to take it
into account. I would not recommend it, though. It is hard to debug.
Peter
J
John Bode
[email protected] said:
This is the wrong question to ask. You don't decide to use a switch
over an if-else chain (or vice-versa) based on performance; for one
thing, the performance difference, if any, will not be consistent from
case to case. Secondly, the intent of the code may be obscured by
using an inappropriate construct, making it harder to read and
understand.
Use a switch when you need a multiway branch based on a single integral
value; use an if-else chain for everything else.
For example, writing
switch(a < b)
{
case 1: /* true */
/* do something */
break;
case 0: /* false */
/* do something else */
break;
}
is silly and a little obfuscatory even if it buys you an extra cycle;
writing
if (a < b)
/* do something */
else
/* do something else */
is clearer and more straightforward. By contrast, even though there's
nothing wrong with writing
if (a == 1)
/* do something */
else if (a == 2)
/* do something else */
else if (a == 3)
/* do yet another thing */
....
this is the kind of situation for which the switch statement was
invented in the first place:
switch(a)
{
case 1: /* do something */
break;
case 2: /* do something else */
break;
case 3: /* do yet another thing */
break;
...
}
B
Bhan
If you know the answer tell me.
Else keep quite.
Else keep quite.
He didn't have enough time to provide context...
There was only 5 minutes left for him to complete his online quiz
and he needed the answers quick!
Besides, the question was asked in the message subject... can't you read?
C
CBFalconer
*** rude top-posting fixed ***
You ARE the rude one. There is still no question to answer.
Message subjects are not necessarily visible when reading a
message.
Bhan said:
You ARE the rude one. There is still no question to answer.
Message subjects are not necessarily visible when reading a
message.
K
Kenny McCormack
The answer is 42.
Quite what?
C
Charles Richmond
I liked the rude guy who told you that you did *not* know anythingCBFalconer said:
about programming. ;-) What a loser!!!
Besides which...the OP's question was implementation dependent and
*not* covered by the C standard.
--
+----------------------------------------------------------------+
| Charles and Francis Richmond It is moral cowardice to leave |
| undone what one perceives right |
| richmond at plano dot net to do. -- Confucius |
+----------------------------------------------------------------+
L
Lawrence Kirby
That depends on what code your compiler decides to generate for each one
and the environment (e.g. processor model) you are running the code on.
Different compilers can generate different code, the same compiler can
generate different code when given different options. The code the
compilergenerates for these will of course depend on the exact tests being
performed in each case, it can also depend on the code around each
statement. Of course for equivalent if-else and switch statements a
compiler could reasonably generate the same output code.
So there is no general answer to your question. For a particualr
combinatoin of code, compiler, compiler options and execution environment
the only reasonable way of geting an answer is to measure both and compare
them. Even that may not be conclusive, the results can depend enormously
on your input data. E.g. in an if-else chain that is translated fairly
directly by the compiler, if a lot of the time the first test matches it
will be fast, if a lot of the time the last test matches or no tests match
it is likely to be slower.
The best approach is usually to express what you want in the clearest way.
It is very often the case that is clearer to read is also easier for the
compiler to optimise.
Lawrence
D
Default User
Bhan said:
*paaaalonk*
Brian