Any ideas on running the float32 object?

Zap-o

I don't understand why I cant get the truncate to work using the Float32 object. Its identical to the Round function as far as I can tell. In wiki the truncation is to the right of the decimal not rounding.

Also what is the Frac, Fmin do? Is there any directions to this object?

Thanks in advance

Mike Green

There's a PDF document included in the Float32Lib download from the Object Exchange. That and any comments in the source code make up the documentation.

"Frac" returns the fractional part of a floating point value.

"Fmin" returns the minimum of two floating point values as shown on page 5 of the PDF file

ElectricAye

I've attached a FloatMath pdf. I'm not sure this is what Mike was talking about, though.

Hope that helps.

ManAtWork

Mike said...
"Fmin" returns the minimum of two floating point values

Hello Mike,

stupid question, but why can't we simly use the ># spin operator? Well, actually the question ist not so stupid, I'll explain... As·mentioned in the wiki article about IEEE-754, floatingpoint numbers have "lexicographical ordering". That means sign and exponent are in the most significat bits. If the floating point values a and b are ordered a<b then their cast-to-integer values (the same bit pattern interpreted as integer) are also ordered a<b (only exceptions are NaN and negative zero).

So in theory we could use all compare and limit-to operators of spin for float values, also. I do this frequently and it seems to work as expected. Are there any objections against doing so?

Cheers

Zap-o

I hate to be the one to point this out, but the directions mention nothing except clock cycles.