I'm sure there is no significant slowdown.
Memcache is some kind of basement for you applications. If you can write in 2 mins a function that will get multiple items from the server (I'm sure you can) than what is the point to implement it in the memcache?

There is more of a slowdown if you are fetching 200+ items from a non-local server, since round trip times add up. Otherwise why would they have added the ability to fetch multiple items at once to the memcache protocol?

199 items fetched in 0.0443150997162 seconds
199 items fetched in 0.473651885986 seconds

First number was using the new pure-php client from cytherianage (http://www.danga.com/memcached/apis.bml, the page appears to be down) with the hashing function modified to use the crc32 method from http://lists.danga.com/pipermail/memcached/2004-July/000713.html

Second number was using the PECL memcache 1.2 client and a PHP class to handle multiple servers, using the same hashing method as above. It fetches multiple items in a simple loop.

Sources:
First class: http://dali.deviantart.com/~chris/memcached-client.php.txt
Second class: http://dali.deviantart.com/~chris/memcache.lib.php.txt
Benchmark: http://dali.deviantart.com/~chris/mctest.php.txt

Ok, I'll take a look at it later a bit.
Feel free to write a patch if you're interested.

This bug has been fixed in CVS.

In case this was a PHP problem, snapshots of the sources are packaged
every three hours; this change will be in the next snapshot. You can
grab the snapshot at http://snaps.php.net/.
 
In case this was a documentation problem, the fix will show up soon at
http://www.php.net/manual/.

In case this was a PHP.net website problem, the change will show
up on the PHP.net site and on the mirror sites in short time.
 
Thank you for the report, and for helping us make PHP better.