[postgis-users] how many min. floating-points?

John Smith wrote:

guys,

Please don't cross-post, especially since nobody on the PostGIS mailing
list answered your previous question.

how many min. floating-points must a server hardware support for
postgresql+postgis? does postgresql+postgis do much floating-point
math to make a difference? can someone give postgresql+postgis
application examples that will require high floating-points?

I don't know what you mean by "floating points". I've never heard it
used in the sense of a countable quantity.

From the manual:

"The data types real and double precision are inexact,
variable-precision numeric types. In practice, these types are usually
implementations of IEEE Standard 754 for Binary Floating-Point
Arithmetic (single and double precision, respectively), to the extent
that the underlying processor, operating system, and compiler support it."

...

"On most platforms, the real type has a range of at least 1E-37 to 1E+37
with a precision of at least 6 decimal digits. The double precision type
typically has a range of around 1E-307 to 1E+308 with a precision of at
least 15 digits. Values that are too large or too small will cause an
error. Rounding might take place if the precision of an input number is
too high. Numbers too close to zero that are not representable as
distinct from zero will cause an underflow error."

I think you would have to have some *very* complicated geography in
order to make PostGIS generate non-negligible floating point-related
imprecision.

Find more information about PostgreSQL and numeric data types at the
following URL.

http://www.postgresql.org/docs/8.3/static/datatype-numeric.html

Colin

John Smith wrote:

guys,
how many min. floating-points must a server hardware support for
postgresql+postgis? does postgresql+postgis do much floating-point
math to make a difference? can someone give postgresql+postgis
application examples that will require high floating-points?
thks, jzs

I do not think there are any limits along these lines in most modern
servers. They all have support for floating point math. If it is not
supported in hardware it is supported in software by the ANSI-C compiler
and libraries.

So the real issue becomes what are you trying to do and what are your
performance requirements? This would be more interesting to us and we
can probably provide better support if you explain what you are trying
to do.

-Steve W

On Thu, Mar 20, 2008 at 02:02:12PM -0400, John Smith wrote:

how many min. floating-points must a server hardware support for
postgresql+postgis? does postgresql+postgis do much floating-point
math to make a difference? can someone give postgresql+postgis
application examples that will require high floating-points?

I think you're referring to "floating point operations per second", not
"floating points"---hence Colin's confusion.

If I understand your question, PG doesn't require any specific
performance level but your application probably does. For example, PG
would be quite happy giving back one row per year if that's all your
processor(s) were capable of. Your users may be a little unhappy with
this though!

What are you doing and what hardware have you tried it on?

Sam

Sam Mason wrote:

On Thu, Mar 20, 2008 at 02:02:12PM -0400, John Smith wrote:

how many min. floating-points must a server hardware support for
postgresql+postgis? does postgresql+postgis do much floating-point
math to make a difference? can someone give postgresql+postgis
application examples that will require high floating-points?

I think you're referring to "floating point operations per second", not
"floating points"---hence Colin's confusion.

Dunno about that. On the PostGIS list, he said:

"i got an old box supporting only 1 floating-point"

Maybe he means an FPU? *boggle*

Just for an interesting comparison, the Casio calculator I've used for
simple stuff for about 15 years runs at about 10 FLOPS. :)

If I understand your question, PG doesn't require any specific
performance level but your application probably does. For example, PG
would be quite happy giving back one row per year if that's all your
processor(s) were capable of. Your users may be a little unhappy with
this though!

I think my calculator could weigh-in around one row per year. ;)

Colin

On 21/03/2008, Colin Wetherbee <cww@denterprises.org> wrote:

Dunno about that. On the PostGIS list, he said:

"i got an old box supporting only 1 floating-point"

Maybe he means an FPU? *boggle*

Maybe floating-point registers on the FPU?

So many options!

Cheers,
Andrej

--
Please don't top post, and don't use HTML e-Mail :} Make your quotes concise.

http://www.american.edu/econ/notes/htmlmail.htm

On Thu, Mar 20, 2008 at 2:16 PM, Colin Wetherbee <cww@denterprises.org> wrote:

John Smith wrote:

guys,

Please don't cross-post, especially since nobody on the PostGIS mailing
list answered your previous question.

On Thu, Mar 20, 2008 at 6:45 PM, Colin Wetherbee <cww@denterprises.org> wrote:

Dunno about that. On the PostGIS list, he said:

please don't cross-post my cross-post. if i wanted to post it to the
postgresql list, i would have ;)
jzs

John Smith wrote:

On Thu, Mar 20, 2008 at 2:16 PM, Colin Wetherbee
<cww@denterprises.org> wrote:

Please don't cross-post, especially since nobody on the PostGIS
mailing list answered your previous question.

please don't cross-post my cross-post. if i wanted to post it to the
postgresql list, i would have ;)

That seems to be quite a silly request, considering you were asking for
assistance on public lists.

Regardless, what did you mean by "floating points"?

We'd like to help, but I don't think we're having much success
interpreting your question.

Some more information about your actual problem would be helpful, too.

Colin

On Fri, Mar 21, 2008 at 10:17 AM, Colin Wetherbee <cww@denterprises.org> wrote:

John Smith wrote:

On Thu, Mar 20, 2008 at 2:16 PM, Colin Wetherbee
<cww@denterprises.org> wrote:

Please don't cross-post, especially since nobody on the PostGIS
mailing list answered your previous question.

please don't cross-post my cross-post. if i wanted to post it to the
postgresql list, i would have ;)

That seems to be quite a silly request, considering you were asking for
assistance on public lists.

no seriously! if i wanted to post it to the postgresql list, i would
have. thanks but no thanks.
jzs

On 22/03/2008, John Smith <jayzee.smith@gmail.com> wrote:

please don't cross-post my cross-post. if i wanted to post it to the
postgresql list, i would have ;)

That seems to be quite a silly request, considering you were asking for
assistance on public lists.

no seriously! if i wanted to post it to the postgresql list, i would
have. thanks but no thanks.
jzs

You did. I can't see what would make you think you hadn't.

Here's the relevant header part from you original message:
--------8<--------8<--------8<--------8<--------SNIP--------8<--------8<--------8<--------8<
Message-ID: <f029597e0803201102s3cb31ec4g4d59368afec9554f@mail.gmail.com>
Date: Thu, 20 Mar 2008 14:02:12 -0400
From: "John Smith" <jayzee.smith@gmail.com>
To: "PostgreSQL General" <pgsql-general@postgresql.org>
Subject: [GENERAL] [postgis-users] how many min. floating-points?
Cc: "PostGIS Users Discussion" <postgis-users@postgis.refractions.net>
--------8<--------8<--------8<--------8<--------SNIP--------8<--------8<--------8<--------8<

Cheers,
Andrej

--
Please don't top post, and don't use HTML e-Mail :} Make your quotes concise.

http://www.american.edu/econ/notes/htmlmail.htm

On Fri, Mar 21, 2008 at 4:24 PM, Andrej Ricnik-Bay
<andrej.groups@gmail.com> wrote:

On 22/03/2008, John Smith <jayzee.smith@gmail.com> wrote:

please don't cross-post my cross-post. if i wanted to post it to the
postgresql list, i would have ;)

That seems to be quite a silly request, considering you were asking for
assistance on public lists.

no seriously! if i wanted to post it to the postgresql list, i would
have. thanks but no thanks.
jzs

You did. I can't see what would make you think you hadn't.

Here's the relevant header part from you original message:

no he dragged my subsequent posts soley to the postgis list to this
list- it is silly for someone who objects to cross-post to cross-post
an objectionable cross-post! have a good weekend, goodbye!
jzs