On 4/21/07, Stut <stuttle@xxxxxxxxx> wrote:
Tijnema ! wrote:
> On 4/21/07, Stut <stuttle@xxxxxxxxx> wrote:
>> Tijnema ! wrote:
>> > There is a difference, echo is slightly faster.
>> > code used for benchmark:
>> > <?
>> > $start = microtime(TRUE);
>> > for ($i=0; $i<100000; ++$i) { print "ABC"; }
>> > echo sprintf("With print ($i): %0.3f\n",microtime(TRUE) - $start);
>> > $start = microtime(TRUE);
>> > for ($i=0; $i<100000; ++$i) { echo "ABC"; }
>> > echo sprintf("With echo ($i): %0.3f\n",microtime(TRUE) - $start);
>> > ?>
>> >
>> > it displays 100000 times ABC, first with the print command, and second
>> > with the echo command. Result:
>> > ABCABCABC<snip>
>> > print (100000): 0.085
>> > ABCABCABC<snip>
>> > echo (100000): 0.076
>> >
>> >
>> > It's not a lot, but since we are displaying data a lot, (most used
>> > function?) it will make a difference in really big scripts.
>>
>> This has been covered before. The difference actually depends on how
>> you're using it, rather than whether you use print or echo. For example,
>> your benchmark shows echo to be slightly faster, but the the following
>> script that I wrote last time this came up shows the opposite. The only
>> difference is that you're outputting a literal whereas I'm printing a
>> variable.
>>
>> http://dev.stut.net/phpspeed/
>>
>> At the end of the day there are more important things to worry about,
>> especially when you're talking in the region of 0.009 seconds per
>> 100,000 calls it's not going to make anywhere near a significant
>> difference to any script you write, even really really big ones scripts.
>>
>> To put it another way, you would need to make 10,000,000 calls for it to
>> extend the runtime of your script by 1 second. Granted you might have a
>> script that calls it 1000 times, meaning 10,000 requests to that script
>> would "waste" 1 second. But unless you're getting twitter-like levels of
>> traffic (they spike at over 11k hits a second) it's not worth worrying
>> about, and I'm guessing (hoping) their devs probably wouldn't care
>> either.
>>
>> Get over it and concentrate on the functionality and usability of your
>> code rather than insignificant details like this.
>>
>> -Stut
>
> Interesting :)
>
> I see there's no big difference between echo and print, but that
> <?=$x?> is faster :)
>
> I've learned (not only from this) that whatever you do in PHP is fast,
> and that you don't need to optimize your code for speed. Unless you're
> hitting 100k+ hits per hour. But even then it would only save you
> maybe one hour per year.
I wouldn't go that far. It is definitely possible to write horribly
inefficient code with PHP. Believe me, I've inherited enough crap code
in my lifetime to testify to that.
My point was simply that you need to look at the numbers from benchmarks
in perspective, and when efficiency is concerned there's almost always
far bigger gains to be made than 0.009 seconds per 100,000 calls to
output something.
-Stut
But what else would you use a lot in your code?
all commonly used things (like while, if, echo, etc) are just (nearly)
as fast as their alternatives (for, print, etc).
Other functions (like file/stream) might be some performance
difference, but you probably use this only a few times in your script.
So there's not a bigger performance difference then when optimizing
echo/print.
Tijnema
--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php
