FloatToStrF returning different values for same data

[Marichele]

Hi,

I have an issue where the value returned from FloatToStrF for the same data input is different for different systems.
Even a hint of issues with floattostrf.

Running the same app on one system (network & pcs) is returning :
output input
1e-130 1e-130, Exponential, 5, 2
1e-18 1e-18, Fixed, 24, 24
1e-10 1e-10, Fixed, 15, 14
1e-10 1e-10, Exponential, 5, 2

Other Network & pcs running the same app with same data:
0 1e-130, Exponential, 5, 2
0 1e-18, Fixed, 24, 24
1e-10 1e-10, Fixed, 15, 14
1e-10 1e-10, Exponential, 5, 2

Past -10 we get zeros.

All pcs are windows xp, various processors that handle xp, good amounts of memory, windows server networks, bandwidth may be different. The other network has extreme security, they guard against anything.

Please any help.

Thanks,
Mary

 

[2ffat]

I would be interest in seeing the source code so I can see which options you are using.


James P. Cottingham
-----------------------------------------
[sup]I'm number 1,229!
I'm number 1,229![/sup]

 

[Marichele]

Actually, I did some more testing and decided it wasn't the FloatToStrF. I think, it may be on of these lines of code.

Result->DetectionLimit = aDataSet->FieldByName("RAW_DETECTION_LIMIT")->AsFloat;

{Processing I was able to eliminate}

aDataSet->FieldByName("DETECTION_LIMIT")->AsFloat = Result->DetectionLimit;

Result->DetectionLimit is a double and the field is an Oracle Number field.
The dataset is coming from TDBGrid where the raw_detection_limit is coming from.

It's strange that only my clients pcs are doing this, every machine I've tried at my office returns the expected value.

We have the clients looking into the issue from their end but I am hoping I can recreate it and therefore guard against it.

Thanks,
Mary