Anybody done any tests for speed, scalability etc?

Does DX Socks performs really better then
any other server components?

Eugene.

--
ICQ: 35477912
MSN: eg...@mts-nn.ru
Yahoo: eugene_goldberg2000
AOL: egold2002

Try the search string:

   borland.public.delphi.internet.winsock: DX versus Indy

or this link:

http://groups.google.com/groups?q=borland.public.delphi.internet.wins...

--
Henrick Hellstr?m,
StreamSec http://www.streamsec.com
Borland Technology Partner
Need to security enhance your Delphi application? Don't settle for
anything less than StrSecII:
http://www.streamsec.com/prod_strsec2.asp

the actual outcome from thae flame war appeared to
be that DXSock outperforms the Indy Server components.
There was no comparisons to other server components.

Grahame
Indy Pit Crew

Grahame Grieve <grah...@kestral.com.au> wrote in news:3DF5DBAB.2070600
@kestral.com.au:

We've never disputed that. However in 98% of the situations, the performance
difference wont matter.

Choose the one that suits your needs better, and is easier to program for
you.

--
Chad Z. Hower (a.k.a. Kudzu) - http://www.hower.org/Kudzu/
      "Programming is an art form that fights back"

  Is Indy useful to you? Send a postcard please!
  http://www.hower.org/kudzu/indypost.html

ELKNews - Get your free copy at http://www.atozedsoftware.com

performance

Chad,

What are the current benchmarks?

No need for a flame.  What are the current numbers?

-dennis

Well, considering that the only benchmark available for DXSocks is from
the DX web site, I don't think you'd find anything conclusive.

DX supporters (and customers) have often claimed here having better
performance than with Indy servers. That's about all you'll be able to
find.

Good luck,
Stephane

Oh, and forget comparing Indy to DxSock or anything else for the time-being.

What are the Indy performance benchmarks?

-dennis

"Allen - Team Indy" <aone...@DONTSPAMspringboardtechnologies.com> wrote in
message news:3df626dc@newsgroups.borland.com...

That is very good to hear.  I see no reason why there are no "official"
benchmarks published.  Full Disclosure is a good thing.

<<<Best thing is to get both, see how easy they fit your needs and make your
own decision.>>

Well, that just sounds like marketing to me.  As a prospective user/customer
there is some basic information I would like up front.  If for no other
reason, then to compare my own performance benchmarks...

-dennis

There is no "official" benchmark because one has never been done.
Full disclosure is great.

Its not - its my advice to you

You might be a potential "customer" for DX, but not for Indy.
Indy is free, and OpenSource - you dont pay anything, and we dont
get paid anything to develop it.

Benchmarks are very subjective, so even more reason for you to do your own,
in your own environment, under your own criteria and conditions, if you need
them.

/A.

Right now, there are no "published" benchmarks - there may be when Indy 10
is ready, but I doubt before. Best thing is to get both, see how easy they
fit your
needs and make your own decision.

/A.

"Allen - Team Indy" <aone...@DONTSPAMspringboardtechnologies.com> wrote in

Not only do you have customers.  We *ALL* have customers.  Except for the
rare hobbyist in these forums, *all* of these tools serve a customer base.

Indy may be free, but the software we develop almost never is.  Thus Indy
isn't free to the end consumer...   That's all I'll say on that...

-dennis

In article <3df61...@newsgroups.borland.com>, "Dennis Landi" <dennis[at]
dennislandi.com> says...
Because we are not in a position to make any objective benchmarks at
all.

First, whatever findings we make could be biased to our own perceptions
and ego involvement in Indy.  That goes the same for DxSock (not that
I'm blasting that component set).  One thing you have to understand is
that often, vendors doing their own benchmarks will tend to get results
favorable to them because of their biases and ego involvement.  Note
that I'm not blasting ego involvement in one's own work in and of itself  
but often, you can not judge a competing product fairly.  Ego
involvement will often compromise any results in research which is why
researchers in many fields are obligated to disclose any conflicts of
interest that can taint their judgment.

Second, we do not have a controlled environment required for tests.  I
say controlled because performance is often effected by external factors
such as network usage at a particular moment (on the Internet, that is
extremely variable), hardware capability, and software setups including
drivers, service packs, and other tasks that run at a particular moment.

Three, we do not have a copy of another vendor's product to test against
and to be frank, I do not want a copy myself because a vendor might
later accuse me of us of stealing their ideas after being exposed to
them.  Such an accusation had been made in the past against one of us.

Four, performance gained in a test application may not translate into
performance in an end-user program.  Other things effect that such as
skill of the programmer and the complexity of the application itself.  
In addition, typical performance will be effected by other factors such
as the server's capability and available bandwidth.  

In and of itself, it is not threatening.  On the other hand, we question
the source of the statistics.  The statics are made by the same person
who had once accused Indy of violating his intellectual property rights
on a public forum.

--
J. Peter Mugaas - Chairperson, Distribution Team, Indy Pit Crew
Internet Direct (Indy) Website - http://www.nevrona.com/Indy
Personal Home Page - http://www.wvnet.edu/~oma00215
If I want to do business with you, I will contact you.  Otherwise, do
not contact me.

Oh good man - you go take everything absolutely literally and not in the
spirit
it was intended. - I intended to say (and was my mistake,  for not being
10000% clear)
that Indy does not have any "paying" customers, because that was the tone
that I
mistakenly picked up from your post.

You are totally right, I am totally wrong, please forgive my PC blindness

/A.

"Dennis Landi" <dennis[at]dennislandi.com> wrote in
news:3df6283c$1@newsgroups.borland.com:

There is nothing thats "not disclosed". The Indy team is very busy with coding
and their real jobs.

If you'd like to run benchmarks, we'd surely assist you in verifying them etc.

Its always best to use your own benchmarks whether they be performance or
suitatiblity. Never trust the benchmarks from a vendor.

--
Chad Z. Hower (a.k.a. Kudzu) - http://www.hower.org/Kudzu/
      "Programming is an art form that fights back"

   Want more Indy stuff? Try the Atozed Indy Portal at
        http://www.atozedsoftware.com/
    * More Free Demos
    * Free Articles
    * Extra Support

ELKNews - Get your free copy at http://www.atozedsoftware.com

"Chad Z. Hower aka Kudzu" <c...@hower.org> wrote in message
news:Xns92E0D1B3D8983cpub@127.0.0.1...
etc.

Thank you.  I am going to convert a very simple delphi app that uses
borland's tClientSocket and tServerSocket components.  I will make the
"sockets" part of this app a plug-in.   That way we can simply use the same
app to benchmark any sockets suite via a runtime package plug-in
architecture.   I will be able to show you the borland sockets version, and
then invite you to come up with an indy (10?) version.

How about it?

This way, all/any vendor can participate without seeing another vendor's
implementation, since what they port is the borland sockets implementation.

Sound good?

-dennis