Offsets Formats

[ark1320]

Some offsets specify the numerical format of the value in the offset, but others do not. Just for example, the offset for Indicated airspeed shows:

02BC   4   IAS: Indicated Air Speed, as knots * 128

So the question that comes up is should this 4 byte (32 bit ) value be read using ipc.readSD(), ipc.readUD(), or ipc.readFLT()? 

Since the format is not specified for this offset (while the format for some other offsets are) is there an implication that any of the above ipc.read instructions will work and the Lua interpreter will make the necessary conversion,  or do you have to read and display the value to see what it looks like, or what? Would appreciate any general guidance for these type of situations.

Thanks,

Al

[Pete Dowson]

1 hour ago, ark1320 said:

02BC   4   IAS: Indicated Air Speed, as knots * 128

So the question that comes up is should this 4 byte (32 bit ) value be read using ipc.readSD(), ipc.readUD(), or ipc.readFLT()? 

If you think you might one day see a negative airspeed maybe you would read it as an SD. But i don't think it is possible in the Sim. So it's best really as a UD.  For positive numbers the only difference bretween a uD and an SD is the maximum value, as the top bit is reserved as the sign.

Floating point and double values are always noted as such. All others are integers in some format or other.

1 hour ago, ark1320 said:

there an implication that any of the above ipc.read instructions will work and the Lua interpreter will make the necessary conversion

No, a plug-in has no idea what the data is supposed to mean, so it can't do anything automatically like that.

Pete

 

 

[ark1320]

30 minutes ago, Pete Dowson said:

Floating point and double values are always noted as such. All others are integers in some format or other.

I assume the above means all single and double floating point values are noted as such, and all others are integers.

Thank you, very helpful!

Al

[Pete Dowson]

Quote

I assume the above means all single and double floating point values are noted as such, and all others are integers.

Yes. "single floating points" are called Floats (FLOAT or FLOAT32) whereas the Doubles are called DOUBLE or even FLOAT64.

Note also that some "integers" are really fixed point values, eg the integer in the high part and fraction in the low part, with the point considered to be at the boundary. But these variations are always described.

This mish-mash is historical. FSUIPC started with FSW95 as a program by Adam Szofran of Microsoft -- I took over with FS98 and later. In the early days FS was designed with memory constraints considered, and data was kept in its most compact form, using units and formats which gave the maximum precision for the least space, at least for all values for which a range or boundary could be determined.

As versions of FS developed this was relaxed, but my ambition with FSUIPC was on-going compatibility with as many of the add-on programs as possible. So, new data formats for new programs but don't stop the older ones working. So FSUIPC still converts most values to their old formats, even though SimConnect supplies (or can be asked to supply) them almost all in standard forms (usually doubles for numerical values, for example). In many cases FSUIPC now supplies values more than once, in different forms.

Pete