CATS : Documention .. Vs .. ERROR

For discussion of how to use the Cycles Research Institute' own Cycles Analysis & Time Series software.

Re: CATS : Documention .. Vs .. ERROR

Postby Steve » 10 Nov 2010, 00:57

Hi Ray,

Thank you for your responses,

However I ask for clarity regarding your statement
" ... the programs for Factor Analysis and Regression both have array sizes limited to n=100 "

Specifically regarding the " limited to n = 100 " portion of your statement

Does your statement mean that
where as above I was attempting an R140 Command to the CATS S/W
that Current V1.0 of CATS S/W puts on the Limit of R100 ?


Now, I do not know which Programming Language you are using to implement your CATS S/W,
However, if using C or more modernly C++ or other Object Oriented Language then
there is NO need to Limit Array Sizes if you properly Code for Dynamic SPACE ALLOCATION

Regards,
Steve
Steve
 
Posts: 16
Joined: 13 Oct 2010, 21:25

Re: CATS : Documention .. Vs .. ERROR

Postby RayTomes » 10 Nov 2010, 22:30

Steve wrote:...
Does your statement mean that
where as above I was attempting an R140 Command to the CATS S/W
that Current V1.0 of CATS S/W puts on the Limit of R100 ?


Now, I do not know which Programming Language you are using to implement your CATS S/W,
However, if using C or more modernly C++ or other Object Oriented Language then
there is NO need to Limit Array Sizes if you properly Code for Dynamic SPACE ALLOCATION

Hi Steve
The mainline of CATS is in FreeBasic. It does have dynamic array allocation which is why these parameters are in our cats.ini file. The mathematical subroutines are in FORTRAN because I have old tested code that I know works for that. It is these FORTRAN routines that have limited array sizes. I don't if this FORTRAN has dynamic allocations. Anyway,we will increase to 500 or something like that. My IT man is busy on a contract so this may not happen for a couple of months.
Regards
Ray
User avatar
RayTomes
 
Posts: 128
Joined: 02 Aug 2010, 23:24
Location: Auckland, New Zealand

Re: CATS : Documention .. Vs .. ERROR

Postby Steve » 12 Nov 2010, 02:38

Hi Ray,

I have noticed that you have
not responded to the questions that
I have previously posted.

I have noticed that you seem to respond to
only " portions of posts containing questions "
but not to the entire content of " the post ".

None-the-Less, your currently so far offered comments as to why
a Regression on 140 Items Fail turns out to be Insufficient !

I therefore assume that you are Guessing, correct me if
my assumption is wrong.

Now, the reason for which I have made the statements in this post are
a) I modified the cats.ini file to be ...
Look at the Picture = http://img715.imageshack.us/img715/3455 ... nov10a.jpg
which content may effect 47 sc Commands via the CATS Command UscAH !

b) I issued the following commands, in sequence, to the Cats V1.0 S/W
1) day
2) UscAH ........ which resulted with 94 items on the Stack
3) F902 # ............ Load the Log of the Data Series
4) R94 ............... results with CRASH

as depicted in http://img842.imageshack.us/img842/9136 ... nov10b.jpg

So, I will re ask about the undocumented restrictions that
demonstratively have been observed regarding STACK Size Limit in the cats.ini file
versus the maximum value of n in an Rn Command.

How does the Stack Size Limit relate to a Limit on the R command ( etc. )

Please be Specific.

Regards,
Steve
Steve
 
Posts: 16
Joined: 13 Oct 2010, 21:25

Re: CATS : Documention .. Vs .. ERROR

Postby RayTomes » 14 Nov 2010, 22:50

Hi Steve

Please don't get stroppy with me.

Steve wrote:...
b) I issued the following commands, in sequence, to the Cats V1.0 S/W
1) day
2) UscAH ........ which resulted with 94 items on the Stack
3) F902 # ............ Load the Log of the Data Series
4) R94 ............... results with CRASH

as depicted in http://img842.imageshack.us/img842/9136 ... nov10b.jpg

So, I will re ask about the undocumented restrictions that
demonstratively have been observed regarding STACK Size Limit in the cats.ini file
versus the maximum value of n in an Rn Command.

How does the Stack Size Limit relate to a Limit on the R command ( etc. )

Please be Specific.

As already stated, the limit of both of both the Rn and Fn commands is n=100. I have looked at the code and that is the limit.

You can set a higher stack size limit. It is not related to those commands. It is controlled by the cats.ini file.

One thing that confuses me is your R94 command. If you have loaded 94 series and then the log of the data before doing R94, then I think there should be 95 things on the stack, not 94. But your image shows 94 things on the stack. Can you confirm whether just prior to issuing the R94 command there are 94 or 95 things on the stack (by using "?" command).

Regards
Ray
User avatar
RayTomes
 
Posts: 128
Joined: 02 Aug 2010, 23:24
Location: Auckland, New Zealand

Re: CATS : Documention .. Vs .. ERROR

Postby Steve » 18 Nov 2010, 23:36

Hi Ray,

With regard to your statement
" One thing that confuses me is your R94 command. If you have loaded 94 series and then the log of the data before doing R94, then I think there should be 95 things on the stack, not 94. But your image shows 94 things on the stack. Can you confirm whether just prior to issuing the R94 command there are 94 or 95 things on the stack (by using "?" command). "

Yes, such confuses me too ...
From the Starting Point on an EMPTY Stack
Each Usc Command places two items onto The Stack
Therefore it follows that 47 individual Usc Commands will
result with 94 Items On The Stack

So there are now 94 items On The Stack, they occupy locations 1 through 94 of The Stack.

Next Step, = Load the " Log of the Data Series " Onto the Stack
via command F902 # ...

The Log of the Data Series now occupies Location 95 of The Stack, specifically
there are now 95 Items On The Stack.

Ray, with specific regard to your question / query
" Can you confirm whether just prior to issuing the R94 command there are 94 or 95 things on the stack "

Indeed, there were 95 Items ( as described above ) On The Stack immediately
before I issued the R94 Command, ... verified by usage of the ? command.

Such verification was done Before
I made my posting of " 12 Nov 2010, 02:38 " above.

Now, the reason that " such confuses me too " is that
I have done other exercises, for Example with the BGMI Index,
where I used 13 Usc Commands followed by
the Loading of the Log of the BGMI Data Series onto the Stack
( the Stack was then occupied by 27 Items ) followed by Commands R26 and then M26
and all seemed to me to work out well.

To me there appears to be " an opportunity for enhancement " to the
undocumented Features of the CATS Software.

In another life ( 1981 - 2002 ) I used to write Software for " The Modern Weapons of War ",
mostly involved with Device Drivers and Device Simulators. In every project I was involved
with it was always that The Biggest Issue was attempting to get The Customer to actually
Define the Testable Characteristics of the PRODUCT they sought.
( I am at least 10 years Out-of-Date as regards current methods of Software )

Now, with regard to your statement
" Please don't get stroppy with me. "

I am only looking for Definitive Characteristics of YOUR Product, how it works
and specific restrictions on its utilization, in-so-much-as the available documentation
really does not ( In My Opinion ) define the Operational Limits of YOUR CATS Software.

Regards,
Steve.
Steve
 
Posts: 16
Joined: 13 Oct 2010, 21:25

Re: CATS : Documention .. Vs .. ERROR

Postby RayTomes » 20 Nov 2010, 22:25

Steve wrote:...
Indeed, there were 95 Items ( as described above ) On The Stack immediately
before I issued the R94 Command, ... verified by usage of the ? command.
...

Hi Steve

Well that eliminates one possible cause of the problem, as I would expect R94 to fail if there were only 94 things on the stack rather than 95.

I will put this on the list of known bugs. My IT man is at present working on a project for someone else, so it might be some time before this is addressed.

Regards
Ray
User avatar
RayTomes
 
Posts: 128
Joined: 02 Aug 2010, 23:24
Location: Auckland, New Zealand

Previous

Return to CATS Software

Who is online

Users browsing this forum: No registered users and 1 guest

cron