NI-XNET supports Signals with a size of max. 52 only. See the NI-XNET Hardware and Software Manual for further information
eg the Signal Property "Number of Bits" explains that pretty accurate
The number of bits the signal uses in the frame payload. IEEE Float numbers are limited to 32 bit or 64 bit.
Integer (signed and unsigned) numbers are limited to 1–52 bits. NI-XNET converts all integers to doubles (64-bit IEEE Float). Integer numbers with more than 52 bits (the size of the mantissa in a 64-bit IEEE Float) cannot be converted exactly to double, and vice versa;
therefore, NI-XNET does not support this. This property is required. If the property does not contain a valid value, and you create an
XNET session that uses this signal, the session returns an error. To ensure that the property contains a valid value, you can do one of the following:
Use a database file (or alias) to create the session.
The file formats require a valid value in the text for this property.
• Set a value in LabVIEW using the property node.
This is needed when you create your own in-memory database (:memory:) rather than
use a file. The property does not contain a default in this case, so you must set a valid
value prior to creating a session.
