Hello and thank you for taking the time from your busy schedule to join in with EZ.
I assume by OPA that you mean op-amps. These devices are intended for a wide variety of amplifier applications whereas the AD8336 is a variable gain amplifier, intended for more limited markets. VGAs amore expensive due to a great deal of additional circuitry to perform the function. Note that the PSRR listed for the AD8336 is a typical specification, at 1 MHz only, and is a general attribute that provided as a reference for the designer to gauge the ability of the device to reject noise that may be contingent to the power supply nets in the application.
I don't know the specifics of OPA specifications, however there may be inconsistencies between the specifications between manufacturers and even between devices from the same manufacturer. The user must examine these specs closely and decide for his or herself the impact on design.
In the case of PSRR, performance is determined with the assumption of no external device by-pass capacitors. This means that users have a great deal to gain by adding generous supply bypassing to their designs. My own rule of thumb is one 0.1uF or greater per device, and one or larger value capacitors for the entire circuit.
I hope this helps,
I have known that you answered the second question, the PSRR of AD8336 is -40dB, to many op-amps, such as ad829, his PSRR is bigger than 0dB, AD8336’s PSRR is smaller than 0dB, that means if the voltage of power changed within 1V, then the output changed within 100V?
Given the relative ages of the two parts and the different development groups it could be a matter of procedure, or what could be called test definitions between the two parts. Any integrated circuit will attenuate power-supply noise at the supply pins, so in my way of thinking that would make for a negative value. The actual value would depend on the test conditions. Unfortunately I don't have a way to refer to test methods used for legacy products.
As you can see the AD829 numbers are very large (94 - 120dB) so I can't think of any test condition that would make noise at the pins 120 dB higher than the reference point, whatever that is. I do wish I had better response for you.