[HELP] error with million value

[diajeng permata]

Dear vensim help/support team,

I am a student. I am using vensim PLE (academic use only) version 6.0b.
I am dealing with million values.

for example, in a level, I inputed 2599796 on "initial value" . But after I close it and open it again, automatically it changed to : 2.5998e+006

How to get complete value ? the "e+" format can not be used for calculation.
I looked at "setting" and "option" menu but i found nothing about this matter.


Thank you very much for your attention

[Administrator]

It should still work ok. What happens when you simulate, it is giving an incorrect value?

[diajeng permata]

actually no incorrect value.. I mean , that's would be an exhausting work If I must convert the results manually from that "e+" format to normal format. Because I need to do the further calculation in excel.

Is there no way to change the setting to always display the complete/normal format? I think there should be a way...

In addition, i don't understand how to read it . For example, if the result is 2.59996e+006 , how to read it in complete/normal format ??

Thank you in advance.

[tomfid]

This is scientific notation, http://en.wikipedia.org/wiki/Scientific_notation

2.53e+6 = 2.53 * 10^6 = 2.53 * 1,000,000 = 2,530,000

Excel understands scientific notation, so there's no problem exporting the results for further calculation.

[SionC]

Hi,

The same thing happens to me (Vensim DSS 6.0a). If you create a constant with a value of 2599796 and then look at the model in text view the value has been changed to 2.5998e+006 rather than 2.599796e+006

All the best

[tomfid]

Again, this is normal behavior - scientific notation.

In DSS, the display of numbers in tables is controllable (right-click the table tool and select "Pretty" or "Scientific" number format). But large numbers in the model equations are always stored in floating point format.

In any case, it would be rare in a dynamic modeling situation for a number to have more than a few significant digits of precision.

[SionC]

Hi Tom,

Thanks. I don't have any problems with scientific notation, and I do not think that this is just an issue of how the numbers are displayed. What is happening is that when you enter a value some of the digits are lost without warning. Another example, if you define a constant of 0.123456789 then it is stored in the text file as 0.1234567...or a value you enter as 123456789.123456789 becomes 1.23457e+008. I understand that this is due to using floating point values and you are right when you say that it is rare in a dynamic modelling situation for a number to have more than a few significant digits of precision. However, it is not impossible and it is something I have had to deal with in the past. To be honest though, the only problem I really have is that this is done without the user being notified the digits have been 'lost', and the first time I came across this it was a pain trying to work out why the results didn't tally with what I expected.

Cheers

[tomfid]

There's a workaround for this if you use Double Precision Vensim. In that case, the file format still stores floats, but (optionally) values can be imported from .cin or spreadsheet files as doubles.