URGENT!!! Can we converse a string of 4 bytes to 32-bit real number?

[plit string]

We have a feedback from our instrument contains a string of 4 bytes representing a 32-bit real number but we can not convert it, can some one help?

Here is a string:     'ffffffee'           

Is there any VI or sample available?

Thank for help  

[sth]

Yes, if you have the string of actual 8 bit data, just separate out each individual unsigned 8 bit number and then combine them using the "Join Numbers" function.  Then join the two 16 bit numbers with another join numbers function.

This function is on the "Numeric Palette" -> "Data Manipulation"  

[MKES]

Check this... and this, maybe is what you need.

[smercurio_fc]

Are you sure you've got the right string? According to IEEE 754 that's not a valid representation in Hex of a 32-bit single-precision number.

[plit string]

Thank for help but I did try & look does not work. I attach my VI so that you can see why I am not right (the number suppose to be -1.xxxxxxxE-3

Can you help one more time?

[plit string]

Yes, it is a right STRING, however it is required doing 2 complement ... because it is a negative number!

I did have another number & it is said to do conversion already (no need to do 2 complement)  bf 53 4c 78   ... but I still in-vain to convert it!

You have any idea?

[plit string]

SORRY gentlements, those 2 strings I was mistaken from the inteprete reading, it shows the equivalent number from the byte  (ie:   '0' ---> 0x03 = 30), so sorry about confuse all of you.

Therefore: 'fffffffee'  is real 4 bytes of  ÿÿÿî

I hope that now we are clear than before! Sorry abot the confusion!

[altenbach]

OK, This is getting silly. Please attach a VI that has the correct string as a diagram constant. Or change your string display to hex and paste the hex representation.

 Do you know if your string is little endian or big endian? ALso tell us what SGL number you expexct as result.

Most likely, all you need is typecast the string. (If it is little endian, use "unflatten from string" and select the correct byte order)

Message Edited by altenbach on 09-09-2008 07:31 PM

[altenbach]

---

plit string wrote:

Yes, it is a right STRING, however it is required doing 2 complement ... because it is a negative number!

---

 Typecasting will take care of this automatically.

---

plit string wrote:

I did have another number & it is said to do conversion already (no need to do 2 complement)  bf 53 4c 78   ... but I still in-vain to convert it!

---

My code would turn a strong of BF53 4C78 (in hex display) into a SGL of -0.825386. Sounds reasonable?

[plit string]

By your explanation, I did setup the instrument with corrected 64-bit reading (since in LabView does not have the 32-bit but DOUBLE instead), so that I use 8 bytes then it works perfectly as your suggested casting

Many thanks