[AccessD] With / End With or not?

[Susan Harkins]

I suppose if you're dealing with tons of data or something -- but I do
disagree that many of today's systems won't respond much faster using With
than not... only in an extreme case. You should do what you want, I'm only
sharing a different perspective that has merit. 

Susan H. 


I don't want to snow on your parade, Susan, but performance *matters!* A
millisecond here and there, accumulated, results in an operation 5 seconds
faster than its ancestor. In many businesses, that lag s crucial. Maybe not
in the realms with which you deal; that's fine. But in other realms
milliseconds are crucial. Take any stock-market app for example: if you are
milliseconds slower than the last transaction you are Yesterday! Nice try,
loser!.

I have written this kind of stuff and the code has to be snappy to the
millisecond. Receive an constant pipe in from Bloomberg and deal with the
items of interest at a few hundred times a second. This is not trivial
Thanks to a colleague, who expert assistance helped me navigate this
terrain, we ultimately made it. It worked, at last, but it took a hundred or
more failed attempts before we finally got there. Blood, Sweat and Tears.

On Mon, Sep 13, 2021 at 4:13 PM Susan Harkins <ssharkins at gmail.com> wrote:

> Performance might have mattered at one time, but with today's systems, 
> I doubt it matters much anymore.
>
> Susan H.
>
>
> I strongly suspect that it's purely  a source code thing and that the 
> compiler will compile both to the same set of instructions using 
> pointers to the appropriate objects.
>
> I find it hard to visualise any other way of it being done.
>
>
>
> On 13 Sep 2021 at 13:59, Arthur Fuller wrote:
>
> > Somewhere I read or heard or just intuited that given a block of 
> > code that frequently references Me, it's faster and better towrap te 
> > the block in a With Me / End With than to explicitly refer to Me. Is 
> > this true? Or is it just to make the code ceaner? Has anyone 
> > bothered to benchmark the difference? I' could easily build a test 
> > case, but if you've already done it, then I would be interested in your
results.
> >
> > --
> > Arthur
> > --
> > AccessD mailing list
> > AccessD at databaseadvisors.com
> > https://databaseadvisors.com/mailman/listinfo/accessd
> > Website: http://www.databaseadvisors.com
> >
>
>
> --
> AccessD mailing list
> AccessD at databaseadvisors.com
> https://databaseadvisors.com/mailman/listinfo/accessd
> Website: http://www.databaseadvisors.com
>
> --
> AccessD mailing list
> AccessD at databaseadvisors.com
> https://databaseadvisors.com/mailman/listinfo/accessd
> Website: http://www.databaseadvisors.com
>


--
Arthur
--
AccessD mailing list
AccessD at databaseadvisors.com
https://databaseadvisors.com/mailman/listinfo/accessd
Website: http://www.databaseadvisors.com