"Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message
news:7E1C1B89-BD09-438B-9D2F-6AFB532A813E@news.elevatesoft.com...

>  If you don't like the performance or don't feel that it is acceptable,
> then you're free to use another product that is more to your liking.

Just another one of them back door guys promoting other products
I guess ?
Rita

Rita,

<< Just another one of them back door guys promoting other products I guess
? >>

I don't know, but there's a heck of a lot more to a product than inserting
100,000+ rows quickly, and I don't really care to spend a lot of time
explaining that again.

--
Tim Young
Elevate Software
www.elevatesoft.com

"Tim Young [Elevate Software]" wrote:

> With all due respect, I have neither the time or the patience to get into a
> long discussion of this benchmark with you right now,

I am sorry that you already feel that way Tim. Our discussion has lasted only by one message. You made some
good suggestions what might affect to ElevateDB performance, and I tried those settings.

>nor do I feel that I need to justify the performance of EDB to you.

I was only looking for the possible causes that could affect to DBISAM/ElevateDb performance. I told earlier
and I am saying it again that my intention was not to offend any of your products. We just have had strange
slowdowns with current DBISAM applications, sought several tips from the newsgroups. And then found that
benchmark code from Binaries group.

We had had talks about amybe upgrading to ElevateDB. Newly re-written code maybe would make the
occasional bottlenecks to go away.  Then we run those tests both with the old engine and then with the new one.

Of course the test did not bring any solution to our current slow downs, but the results made me wonder why
there had been almost no discussions about those results. Then I found myself participating to this "DBISAM 3
or 4" thread. And now it looks like this discussion turned too severe, and was closed from the vendor's side.

I just wonder if all DB Engine performance related questions have been banned from everyone, or is it this time
me only

> because I knew that it would result in such a response.  

I am truly sorry if I did say something offending. I just tested the improvements you told, and then reported the
resulting numbers back.

> If you don't like the performance or don't feel that it is acceptable, then you're
> free to use another product that is more to your liking.

I know that DBISAM/ElevateDB are important to you and your business. Everyone wants to protect their living.

But you also should realise that I have not invented these performance questions myself.  They come from our
customers, our DB apps having strange slow downs. Some simply looking queries seem to take too long, no
matter how you optimize them etc.

My intention with that easily repeatable benchmark was only to find some simple measurements. And common
expressions how one could talk about these numbers and matters.

Everyone has to solve problems within their applications in some way. Now it looks like this road is closed,
blocked, and one has to continue looking either something else or somewhere else.

At least I got some response. I looks like also some others have had the these kind of questions in their mind
earlier. Just now it was not good time to talk about them.

I wonder if there will ever be good time for them. Thanks for everyone participating to this thread.
Stan

Stan,

<< I was only looking for the possible causes that could affect to
DBISAM/ElevateDb performance. I told earlier
and I am saying it again that my intention was not to offend any of your
products. We just have had strange
slowdowns with current DBISAM applications, sought several tips from the
newsgroups. And then found that
benchmark code from Binaries group.  >>

And what did you possibly expect to glean from the benchmark regarding your
issue ?  Furthermore, why didn't you just email us or ask specifically for
support on your issue instead of going on about some very old benchmark ?

This was your first post on this issue:

"I just mean that old 3.x version seem to beat those new flag ships in
benchamrk tests.  I especially hate to see
NexusDB to beat all these "our" database versions in speed."

That sure doesn't sound like a support issue to me.

I let it go on for a while, but as soon as it starts resulting in me being
forced to discuss a bulk insert benchmark instead of doing something
productive, then it's over.  If you have a specific support issue, then post
it or send us an email at support@elevatesoft.com and we'll be happy to help
you out.

--
Tim Young
Elevate Software
www.elevatesoft.com

"Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message
news:A7831A81-0A06-4EBD-98BC-664AEB63CDAC@news.elevatesoft.com...
>
> "I just mean that old 3.x version seem to beat those new flag ships in
> benchamrk tests.  I especially hate to see
> NexusDB to beat all these "our" database versions in speed."
>
> That sure doesn't sound like a support issue to me.

It's not up to you, I think, to define what is or is not a valid support
issue. All of us DBISAM users have to justify why we use DBISAM instead of
one of the "big ones". If somebody is 1) considering an upgrade, never a
trivial issue and 2) concerned about speed and 3) finds out that - rightly
or wrongly - his tests indicate that the new product is slower than the old
one, or the competition, then he has a perfect right to be concerned. After
all he has to explain to his company why he wants to spend time and money
upgrading to EDB instead of Nexus.

I remember at least one other issue in which you threw a tantrum at me
instead of addressing the issue. I think you need to be careful with that
temper business. This is not a personality or feelings thing. We are in
business, OK? And furthermore, in the software busines. If we of all people
can't address issues with logic instead of feelings, we have a problem.

>
> I let it go on for a while, but as soon as it starts resulting in me being
> forced to discuss a bulk insert benchmark instead of doing something
> productive, then it's over.

Why is it over? What if Stan and I wanted to continue and discuss this issue
further? What if other users had good insights? That is what a newsgroup is
for.

Robert

"Robert" <ngsemail2005withoutthis@yahoo.com.ar> wrote in message
news:CA7963CD-7C6A-4ED8-A1CB-DE12312EA686@news.elevatesoft.com...
>
>
> I remember at least one other issue in which you threw a tantrum at me
> instead of addressing the issue. I think you need to be careful with that
> temper business. This is not a personality or feelings thing. We are in
> business, OK? And furthermore, in the software busines. If we of all
> people can't address issues with logic instead of feelings, we have a
> problem.
>

I think you have been working with the courts so much you
are sounding like a judge
Rita

"Rita" wrote:

>I think you have been working with the courts so much you
>are sounding like a judge

While Tim has closed the further conversation I do not want to continue this thread too
long. Yet while Robert was willing to speak up for my thoughts, I also want to put up a
few words to comment his sayings as I see there quite similar thoughts that we have had in
our minds.

"Robert" wrote:
> After all he has to explain to his company why he wants to spend time and money
> upgrading to EDB instead of Nexus.

Occasional slow downs was our original problem. To clear this thing, we currently have *no
plans* going or upgrading especially to Nexus. That engine name just happened to be among
in that benchmark test, and it  was able to run without needing any extra DLL-files like
Advantage, and BDE files I already had.

I do not see how that benchmark was "very old benchmark" as I could see inside the sources
what its does, and I re-compiled it yesterday. That benchmark also seems to be the only
one anyone has. I will gladly run any other benchmarks there might be.

Of course we might have to move elsewhere if numerical facts and also feelings say you
have to go there. Personal feelings aren't very good right now.

But I also understand Elevate's thoughts and position with this matter. Even if you have
good references (=sources) to follow, no big improvements can appear overnight, it may
take months. I would gladly have given a full year or even more, if any of my performance
concerns had got any response.  The added new engine features are nice, but it alone can
not compensate the pure performance needs.

It may be after all very relatively small changes that would make the difference. That
means if there is any need for that in the first hand. It _may be_ that DBISAM/ElevateDB
really beats those other engines in multi user environment, hands down. But those kind of
test measurements will now never appear, at least not within this closed thread.

If we have to start using some other database with some of our installations, it may be
something else than any of the engines mentioned in this thread. That's why I thought I'll
add a few more engines to the benchmark, to get ourselves yet more numbers to think about.

"Robert" wrote:
> If we of all people can't address issues with logic instead of feelings,
> we have a problem.

For Tim I have to say that I also have my own temper, and it is not always so good. I have
not directly shouted to clients, but I have sometimes been a way more short worded and
impatient than necessary.

Also English is not my native, and I may have to use dictionary when trying to find more
"colorful" expressions. I now better see my opening message at the original humorous and
cheerfuls thread situation. It was not good opening, when you look it against the latter
thread development, so my apologies for that.
Stan

Robert,

<< It's not up to you, I think, to define what is or is not a valid support
issue. >>

These are our newsgroups and we pay for them, and I manage the support
requests, so yes, it *is* up to me.

This thread is done.  I will not discuss these types of benchmarks anymore
because they are misleading and provide only a small slice of information
about a product.

If Stan has a performance issue, he can post a support request at any time.

--
Tim Young
Elevate Software
www.elevatesoft.com

No.  We customers pay for them.  

I have a support question.  Why is EDB insert speed so slow?  How can it be sped up to
match the competition?

A long paying customer,
Rob

"Tim Young [Elevate Software]" wrote:

These are our newsgroups and we pay for them, and I manage the support
requests, so yes, it *is* up to me.

This thread is done.  I will not discuss these types of benchmarks anymore
because they are misleading and provide only a small slice of information
about a product.

If Stan has a performance issue, he can post a support request at any time.

--
Tim Young
Elevate Software
www.elevatesoft.com

Rob


Some suggestions

1. If this is a genuine support question start a new thread - Tim has already said he won't respond to this one
2. Quote specific examples where the speed is to slow for your needs (note I do not say where its slower than other db engines)
3. Try to avoid direct confrontational statements unless all you are looking for is an argument

Finally, a question back:

<<Why is EDB insert speed so slow?  How can it be sped up to
match the competition?>>

How slow? what competition? Have you developed benchmarks to prove this assertion for each of the engines?

Roy Lambert [Team Elevate]

Also a long paying customer (sine 2001 I think)