On 24.5.2004, at 13:47, Joseph J. Strout wrote:
Yes, but "much slower" is not the same thing as "too slow." Is there
any real-world app where the time spent inside REALLoadObjectMethod
accounts for even 1% of the CPU time during some process?
If not, then even eliminating this method entirely could not speed
that part of the app up by as much as 1%, and it would be incorrect
for us to make the framework code larger and more complex for no
perceivable benefit.
Look at Ruslan's case, he needs to talk to a REALbasic variant in a
tight loop for Recordsets,
I would say that was a very speed critical thing, and I would say it
accounts for more than 1% speed increase.
I put such in the Grid rendering engine, and scrolling got way slow,
reverting back to unpublished entry points made it fast again.
The thing is you cant really say that this unneeded slowdown is
acceptable just because you cant think of cases where such speed is
needed.
If none of us were concerned about speed then I think we would all be
using Java and not REALbasic.
--
______________________________________________________________________
Björn Eiríksson bjorn at einhugur dot com
Einhugur Software
http://www.einhugur.com/
______________________________________________________________________
Einhugur Software has sold its products in 43 countries world wide.
______________________________________________________________________
For support: support at einhugur dot com
To post on the maillist: einhugur at yahoogroups dot com
_______________________________________________
Unsubscribe or switch delivery mode:
<http://support.realsoftware.com/listmanager/>
Search the archives of this list here:
<http://support.realsoftware.com/listarchives/lists.html>
|