Note that you can set the channel labels to show the FFT window function in screenshots....
Also, while we wait for Rigol to add basic averaging features to their FFT, one can use pyvisa to acquire traces from the scope and display the avg result with a few lines of python code. Here is a minimalist working example:
import pyvisa as visa
import matplotlib.pyplot as plt
import numpy as np
scopedev = 'TCPIP0::192.168.0.10::5555::SOCKET' #four 5's
nSamples = 100
scope = visa.ResourceManager().open_resource(scopedev)
scope.read_termination = scope.write_termination = '\n'
print("Scope: ", scope.query("*IDN?"))
scope.write(":WAV:SOURCE MATH1;:WAV:FORM ASCii;:WAV:MODE NORMAL")
fstart = float(scope.query(":MATH1:FFT:FREQuency:START?"))
fend = float(scope.query(":MATH1:FFT:FREQuency:END?"))
X = np.linspace(fstart, fend, 1000, endpoint=True)/1e6
fftData =[]
for i in range(0,nSamples):
fftData += [np.array(scope.query(":WAV:DATA?").split(","),dtype=float)]
avgFFTdata = np.mean(np.array(fftData),axis=0)
plt.plot(X,avgFFTdata)
plt.grid(True, which="both")
plt.show()