I am trying to implement a GUI on a Raspberry Pi which plots real time data coming in from a TCP client with the functionality of panning and zooming as well. I have heard all the rage about PyQtGraph being faster than Matplotlib and would like to reimplement the Gui I'd previously made using Matplotlib and Gtk using PyQtGraph and PyQt5. I have benchmarked the code but would like clarification on the results seeing as I can blit the Matplotlib plot to dramatically reduce the plot time. Below is the code I used for benchmarking:

# PyQtGraph
import pyqtgraph as pg
import numpy as np
app = pg.mkQApp()
x = []
y = []

for i in range(1000000):
    x.append(i)
    y.append(1)

view = pg.GraphicsLayoutWidget()
view.show()
w1 = view.addPlot()
w1.disableAutoRange()
start = pg.ptime.time()
w1.plot(x, y)
w1.autoRange() # only after plots are added
end = pg.ptime.time()
print("Plot time: %0.2f sec" % (end-start))
app.exec_()

and

# Matplotlib
import matplotlib.pyplot as plt
import time
x = []
y = []

for i in range(1000000):
    x.append(i)
    y.append(1)

fig, ax = plt.subplots(nrows=1)
(line,) = ax.plot([], [], "b")

plt.show(block=False)

start = time.time()
line.set_data(x, y)
ax.relim() 
ax.autoscale_view(True,True,True) 
fig.canvas.draw()
fig.canvas.flush_events()
end = time.time()

print(end-start)

# In order to see if it has plotted
time.sleep(10)

For 1 000 000 points, I got 2.54 s to plot using PyQtGraph and 6.97 s to plot using Matplotlib on my Raspberry Pi. Is this the expected speed up or is there something else I'm not doing? Thank you.