Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

New FF2 Transports

1 view
Skip to first unread message

Rodrigo Gómez

unread,
Nov 26, 2002, 7:14:17 PM11/26/02
to
Hi Paul.

I'm very interested in the speed gains, as I guess a lot of people will
be.

The only "problem": is it factible to have a C++Builder version? 99.9%
of my programs are developed in C++Builder, and changing them to Delphi is
not really an option <bg>. (unless of course the speed gains were more than
7x... :-) )

Thanks, and regards

Rodrigo Gómez


"Paul Motyer [TPX]" <nos...@no.spam.com> escribió en el mensaje
news:d5Bn3QZl...@tpsmail01.turbopower.net...
> I'm beta testing my new FF2 transports. Currently only for Delphi 7
> and FF 2.12 - for URL see the signature
>
> --
> ,-._|\ Paul Motyer [TPX]
> / Oz \ SoftStuff P/L
> \_,--._/ [paul.motyer at tpx.turbopower dot com]
> v [http://members.optushome.com.au/paulmotyer]
>
>


Rodrigo Gómez

unread,
Nov 26, 2002, 7:57:36 PM11/26/02
to
>
> Sorry, I don't have C++B.
>

If I may be of some use, please let me know. I have C++Builder 4 &
C++Builder 5. I'm planning to buy C++Builder 6, in the next months. I'm
no-way a FF guru, but I can certanly compile and try to hack things to be
compilable under C++builder (already did that another several times, by
neccesity)

And I was joking about speed. It is great, but not really a reason I can
port my stuff to Delphi.

That's why I want your new transports to be usable under C++Builder. Of
course I'm also interested in the security advantages of your work, but it
is easy to overlook that having a 15.3:3.59, 28.9:8.9 differences... :-)

By the way, I assume that you'r work is going to be in the FF3 stuff (just
look at the screenshots!) <vbg>. Of course you can't talk about that. :-)

Regards,

Rodrigo Gómez


> Besides speed, there is also security. The TCP transport can use
> Blowfish (or some other) encryption
> Re speed: the following times are the time to scan through a table
> with First; while not Eof do Next; - all with the client & server on
> the same machine
>
> ------------
> for a 135 Mb table with 1,557 record length & 48,000 records:
> No Transport/Embedded Engine Time: 1.57859169251958 sec
> Shared Mem Time: 3.59144472272314 sec
> Blocking TCP Time: 8.91269225557997 sec
> SUP Time: 15.3247580348899 sec
> TCP Time: 28.9904596813282 sec
>
> Relative to Legacy TCP speed = 1.0
> SharedMem: 13.6 times faster than Legacy TCP
> Blocking TCP: 3.7 times faster than Legacy TCP
> Legacy SUP: 2.0 times faster than Legacy TCP
> ------------
> For a 3 Mb table with a 28 byte record length & 29,000 records:
> No Transport/Embedded Engine Time: 0.408092648646686 sec
> Shared Mem Time: 1.37741444792564 sec
> Blocking TCP Time: 3.67035112004459 sec
> SUP Time: 6.13532781400988 sec
> TCP Time: 13.9690094690806 sec
>
> Relative to Legacy TCP speed = 1.0
> SharedMem: 14.0 times faster than Legacy TCP
> Blocking TCP: 4.2 times faster than Legacy TCP
> Legacy SUP: 2.4 times faster than Legacy TCP
> ------------

Paul Motyer [TPX]

unread,
Nov 26, 2002, 8:22:12 PM11/26/02
to
"Rodrigo Gómez" wrote

> By the way

oops, missed that one - thanks for the warning - fixed

Paul Motyer [TPX]

unread,
Nov 26, 2002, 7:48:43 PM11/26/02
to
"Rodrigo Gómez" wrote

> a C++Builder version?

Sorry, I don't have C++B.

Besides speed, there is also security. The TCP transport can use

--

William E. Mahaffey

unread,
Nov 27, 2002, 1:34:52 PM11/27/02
to
Rodrigo and Paul,

I too am a C++ Builder user and have C++ Builder 6. If I can help, I will
be pleased to do so.

William Mahaffey

"Rodrigo Gómez" <rgomez...@THIStsradio.com.mx> wrote in message
news:AZ6f9$alCHA...@tpsmail01.turbopower.net...

0 new messages