On May 6, 2009, at 6:49 PM, Josh Berkus wrote:
On 5/6/09 9:02 AM, Brian Aker wrote:
Hi!
Here is another question though... one may be faster, but which one
can
we maintain? Processor speed is going up, one memory fetch can
equal how
many operations?
If we make it so that only a few people can look at the code... or
that
the few that can have to spend hours on it... fast is required, but
what
happens to maintenance?
What *is* the real-world performance difference? Using a custom
library understood by only one person in order to get a 2%
improvement is foolish, but if it makes a *large* difference, it
might be worth it. In that case, though, Mats' code needs to be
incorporated into Drizzle and well-documented (and not just "magic",
even if Mats *is* way smarter than me).
Actually I believe 2% does make a difference!
The problem is, it all adds up!
If we make the same decision 10 times we have a 20% slowdown.
So I don't think that we can compromise in speed at all. The only
possibility is to document the code, and keep it modular, to make it
as easy to maintain as possible.
--
Paul McCullagh
PrimeBase Technologies
www.primebase.org
www.blobstreaming.org
pbxt.blogspot.com
_______________________________________________
Mailing list: https://launchpad.net/~drizzle-discuss
Post to : [email protected]
Unsubscribe : https://launchpad.net/~drizzle-discuss
More help : https://help.launchpad.net/ListHelp
