Visualization of spin-systems in QDYN¶
The nlevel_spin.f90 module offers a variety of tools for visualizing N-level spin states. Among these, you can visualize your state via the Wigner-spin function or the Husimi-Q-function. But it is also possible to trace your spin-coherent states over time.
As an example, we demonstrate these visualization tools on a example system and give some ideas on how to present the results in this tutorial.
First, we will show how to visualize spin-systems in QDYN via the spin-wigner function. As model system we will choose the Bose-Hubbard model from Kallush et al 2014 New J. Phys. 16 015008, (eq. (34)):
\begin{equation} \hat{H} = -2\Delta\hat{J}_x + U\hat{J}^2_z + 2u(t)\hat{J}_z, \end{equation}
where \(\Delta\) and \(U\) are postive parameters, and u(t) is the control pulse.
Note: We implement but don’t use the control pulse in this example! Feel free to play around with it!
In the following we will first set up the script.
[1]:
import copy
from pathlib import Path
import numpy as np
import qutip
import matplotlib.pyplot as plt
import colorsys
from mpl_toolkits.mplot3d import Axes3D
import qdyn.model
import qdyn.pulse
!make clean
[2]:
j=2*7+1
Implementing the model¶
First, we define a function which sets the Hamiltonian
[3]:
def get_Ham(
j=2, delta=15, U=None # use default value from paper
):
if U is None:
U = 2.0 * delta / j
u = lambda t, args: 0
Ham_0 = -2.0 * delta * qutip.operators.spin_Jx(j) + U * (
qutip.operators.spin_Jz(j) ** 2.0
)
Ham_1 = 2.0 * qutip.operators.spin_Jz(j)
return [Ham_0, [Ham_1, np.vectorize(u)]]
Then, we define a function which writes out all the necessary data to the runfolder which we choose to be “rf” as default
[4]:
def qdyn_model(
Ham,
qdyn_tlist,
psi0,
results_folder,
print_time_needed = True,
picture_dims=[500,250], # default in the program
runfolder="rf",
prop_method="cheby",
**user_kwargs,
):
""" Create qdyn-model
"""
dt = (qdyn_tlist[0]) / (qdyn_tlist[1] - 1)
tgrid = np.linspace(
float(dt / 2),
float(qdyn_tlist[0] - dt / 2),
qdyn_tlist[1] - 1,
dtype=np.float64,
)
model = qdyn.model.LevelModel()
for H in Ham:
if isinstance(H, list):
model.add_ham(
H[0],
pulse=qdyn.pulse.Pulse(
tgrid,
amplitude=H[1](tgrid, None),
time_unit="iu",
ampl_unit="iu",
config_attribs={
'oct_shape': 'const',
},
),
op_unit="iu",
)
else:
model.add_ham(H, op_unit="iu")
model.set_propagation(
qdyn_tlist[0], qdyn_tlist[1], time_unit="iu", prop_method=prop_method
)
model.add_state(psi0, 'initial')
user_variables = {}
user_variables["phi_pixels"] = picture_dims[0]
user_variables["theta_pixels"] = picture_dims[1]
user_variables["results_folder"] = str(Path(runfolder)/results_folder)
user_variables["print_time_needed"] = print_time_needed
user_variables.update(user_kwargs)
model.user_data = user_variables
model.write_to_runfolder(str(runfolder))
Path(user_variables["results_folder"]).mkdir(exist_ok=True)
Trajectory on the Bloch sphere¶
Set up everything¶
For the propagation we use a spin-coherent state at the north pole, \(\theta=0\), \(\phi=0\)
[5]:
psi0 = qutip.states.spin_coherent(j, 0, 0)
The tempral grid for qdyn is given as qdyn_tlist, which is a tuple (T, Nt). We will then propagate on a grid from 0 to T, with Nt timesteps
[6]:
qdyn_tlist = ( .1 , 100 ) #= ( T , Nt )
And finally we set up the model by using the routine above. We use a Hamiltonian without the squeezing term, i.e. U = 0.
[7]:
results_folder = Path("results_btraj")
[8]:
H = get_Ham(j,U=0)
qdyn_model(
H, qdyn_tlist, psi0,
results_folder=results_folder,
plot_qfct = False,
plot_wigner = False
)
Run the script¶
[9]:
!make
ifort -O2 -openmp -vec-report-0 -par-report-0 -openmp-report-0 -mkl=sequential -I`pwd` -I../../../../ -I/home/users/0000/uk060352/local/include -L../../../../ -L/home/users/0000/uk060352/local/lib -c -o prop_spin_sys.o prop_spin_sys.f90
ifort -O2 -openmp -vec-report-0 -par-report-0 -openmp-report-0 -mkl=sequential -I`pwd` -I../../../../ -I/home/users/0000/uk060352/local/include -L../../../../ -L/home/users/0000/uk060352/local/lib -o prop_spin_sys prop_spin_sys.o -lqdyn -ldfftpack -llbfgsb -larpack -lwigxjpf -lfastwigxj
[10]:
!./prop_spin_sys
*** Read config file rf/config ***
*** Done reading config file ***
*** Initializing system ***
*** Initializing grid ***
No explicit grid found in config file! Initializing grid as one dimensional 1x1 fake grid
Initializing grid as 1D cartesian grid without mapping
Number of grid points was set to nr = 1 for dimension dim = 1! Initializing dimension 1 as a 1x1 fake-grid...
*** Initializing pulses ***
1 pulses in config file
Initializing pulse 1
setting pulse shape as constant 1.00 in [ 0.000E+00, 1.000E-01]
*** Initializing dynamical generator ***
*** Done with initialization ***
Obtaining the expectation values in the Bloch sphere: 1.553392410278320E-002
s
Plot data in a graph¶
First, we plot the expectation values in a simple graph. Here, we see the temporal evolution of the expectation values \(\langle j_i \rangle\) for \(i=x,y,x\), and the spherical coordinated \(r\),\(\theta\) and \(\phi\). Note that the description of a n-level system by means of it’s expected position on the Bloch sphere is only meaningful, if the state is (close to) a spin-coherent state. The variable ovlp shows the overlap between the propagated state and the closest
SCS which is, in this example, \(1\). That means that we follow the dynamics of a SCS on the surface of a generalised Bloch sphere.
[11]:
def plot_exp_values(dat_path, plot_path, plot=True):
results_dat = sorted(dat_path.iterdir())
r = np.loadtxt(results_dat[0])
fig, ax = plt.subplots()
leg = ['jx','jy','jz','r','theta','phi','ovlp']
for i in range(1,len(r[0,:])):
ax.plot(r[:,0],r[:,i],label=leg[i-1])
ax.legend(ncol=7,mode="expand")
ax.set_xlabel("ti")
Path(plot_path).mkdir(parents=True, exist_ok=True)
plt.savefig(Path(plot_path)/"bloch_exp_values.png", dpi=200)
if plot:
plt.show()
plt.close(fig)
plot_exp_values("rf" / results_folder, "pictures")
The data is easier to understand if we look at the dynamics on the Bloch sphere. The blue dots show the trajectory of the center of a SCS on the Bloch sphere.
[12]:
def plot_bloch_traj(dat_path, plot_path, plot=True):
results_dat = sorted(dat_path.iterdir())
r = np.loadtxt(results_dat[0])
b = qutip.Bloch()
b.view = [60,210]
b.add_vectors([r[0,1],r[0,2],r[0,3]]) # initial state, green
b.add_vectors([r[-1,1],r[-1,2],r[-1,3]]) # final state, orange
b.add_points([r[:,1],r[:,2],r[:,3]])
b.show()
Path(plot_path).mkdir(parents=True, exist_ok=True)
plt.savefig(Path(plot_path)/"bloch_traj.png", dpi=200)
plot_bloch_traj("rf" / results_folder, "pictures")
Note: This cellfails for some versions of qutip and matplotlib. You can solve this by updating both (for example via pip install qutip matplotlib --upgrade).
Spin-Wigner and Q-function¶
If the state is not (close to) a SCS, the mean position of the state on the generalised Bloch sphere is not sufficient anymore and we move on to two more suitable visualisations: the Wigner function and the Q-function.
Set up everything¶
For the propagation we use a spin-coherent state at \(\theta=\pi/2\), \(\phi=\pi\)
[13]:
psi0 = qutip.states.spin_coherent(j, np.pi/2, np.pi)
The tempral grid for qdyn is given as qdyn_tlist, which is a tuple (T, Nt). We will then propagate on a grid from 0 to T, with Nt timesteps
[14]:
qdyn_tlist = ( .1 , 10 ) #= ( T , Nt )
And finally we set up the model by using the routine above. We use the default values for the Hamiltonian, i.e. the one with the squeezing term \(U\).
[15]:
results_folder = Path("results_qfct_wigner")
N_theta = 150
N_phi = 75
[16]:
H = get_Ham(j)
qdyn_model(
H, qdyn_tlist, psi0,
results_folder=results_folder,
picture_dims = (N_theta, N_phi),
exp_bloch = False
)
Run the script¶
[17]:
!make
make: 'prop_spin_sys' is up to date.
[18]:
!./prop_spin_sys
*** Read config file rf/config ***
*** Done reading config file ***
*** Initializing system ***
*** Initializing grid ***
No explicit grid found in config file! Initializing grid as one dimensional 1x1 fake grid
Initializing grid as 1D cartesian grid without mapping
Number of grid points was set to nr = 1 for dimension dim = 1! Initializing dimension 1 as a 1x1 fake-grid...
*** Initializing pulses ***
1 pulses in config file
Initializing pulse 1
setting pulse shape as constant 1.00 in [ 0.000E+00, 1.000E-01]
*** Initializing dynamical generator ***
*** Done with initialization ***
Initialising wigner symbols on 32 threads:
Time needed for wigner 5.894494056701660E-002
Time needed for wigner 3.347587585449219E-002
Time needed for wigner 2.896595001220703E-002
Time needed for wigner 2.896595001220703E-002
Time needed for wigner 1.561212539672852E-002
Time needed for wigner 1.571798324584961E-002
Time needed for wigner 1.569819450378418E-002
Time needed for wigner 1.557993888854980E-002
Time needed for wigner 1.560711860656738E-002
Time needed for wigner 1.557397842407227E-002
Calculating and writing the spin-wigner function took: 0.274631977081299
s
Time needed for q-fct 0.153646945953369
Time needed for q-fct 0.154102802276611
Time needed for q-fct 0.153930902481079
Time needed for q-fct 0.153733015060425
Time needed for q-fct 0.153908014297485
Time needed for q-fct 0.153897047042847
Time needed for q-fct 0.153969049453735
Time needed for q-fct 0.154065847396851
Time needed for q-fct 0.189017772674561
Time needed for q-fct 0.195749044418335
Calculating and writing the q-function took: 1.63815498352051 s
Load data¶
First of all, we set a variable for the results folder, which we set earlier
[19]:
res_folder = "rf" / results_folder
results_dat = sorted(res_folder.iterdir())
The resulting data can now be loaded from the files by using numpy’s loadtxt function …
[20]:
r = np.loadtxt(results_dat[0])
… and plotted:
[21]:
fig, ax = plt.subplots()
ax.pcolormesh(r)
ax.axes.set_aspect('equal')
Plotting everything¶
We will now load all the saved files and plot them. To save some work, we will first write a function, that first loads all the data and then plots it:
[22]:
def plot_surface(dat_path, plot_path, plot_inline=True):
results_dat = sorted(dat_path.iterdir())
results = []
for dat_file in results_dat:
results.append(np.loadtxt(dat_file))
# get the number of files, the first half of them are the Q-functions and the other half the Wigner functions
num_values = len(results)
mid = int(num_values/2)-1
# get the maximum value of all results for proper scaling
# (it's a different one for the Q-function and the Wigner function)
max_values = [np.amax(abs(np.array(results[0:mid]))),\
np.amax(abs(np.array(results)[mid+1,:]))]
plot_path = Path(plot_path)
if not plot_path.exists():
plot_path.mkdir()
for idx, res in enumerate(results):
if idx <= mid:
mode = 0
idy = idx # index of the plot, starting from 0
filename = "qfct_"+f"{idy:0>5}.png"
else:
mode = 1
idy = idx - mid - 1 # index of the plot, starting from 0
filename = "wigner_"+f"{idy:0>5}.png"
fig, ax = plt.subplots()
phi, theta = np.meshgrid(np.linspace(0,2*np.pi,res.shape[1]),np.linspace(0,1*np.pi,res.shape[0]))
ax.pcolormesh(phi, theta, res, vmin = -max_values[mode], vmax = max_values[mode], cmap="bwr")
ax.set_xlabel(r"$\phi$", usetex=True)
ax.set_ylabel(r"$\theta$", usetex=True)
ax.axes.set_aspect('equal')
plt.savefig(Path(plot_path)/filename, dpi=200)
if plot_inline:
print(Path(plot_path)/filename)
plt.show()
plt.close(fig)
[23]:
%%capture
# `%%capture` suppresses the output and can be removed while playing around ;-)
plot_surface("rf" / results_folder, "pictures", plot_inline=True)
Plotting on a sphere¶
Instead of plotting a 2D plot, we can also visualise the state on a 3D sphere:
[24]:
def plot_sphere(dat_path, plot_path, plot_inline=True, plot_ids = []):
results_dat = sorted(dat_path.iterdir())
results = []
for dat_file in results_dat:
results.append(np.loadtxt(dat_file))
# get the number of files, half of them are the Q-function and half the Wigner function
num_values = len(results)
mid = int(num_values/2)-1
# get the maximum value of all results for proper scaling
# (it's a different one for the Q-function and the Wigner function)
max_values = [np.amax(abs(np.array(results[0:mid]))),\
np.amax(abs(np.array(results)[mid+1,:]))]
plot_path = Path(plot_path)
if not plot_path.exists():
plot_path.mkdir()
for idx, res in enumerate(results):
# If plot_id is given, we only plot the file with index `plot_id`
if plot_ids:
if (idx not in plot_ids):
continue
if idx <= mid:
# Q-function
mode = 0
idy = idx # index of the plot, starting from 0
filename = "sph_qfct_"+f"{idy:0>5}.png"
# normalise the Q-function to [0,1] using `scale`
scale = max_values[mode] # because the values go from 0 to max_value
shift = 0
mycmap = copy.copy(plt.get_cmap("jet")) #non-divergent colormap
else:
# Wigner
mode = 1
idy = idx - mid - 1 # index of the plot, starting from 0
filename = "sph_wigner_"+f"{idy:0>5}.png"
# normalise the Wigner function to [0,1] using `scale` and `shift`
scale = 2*max_values[mode]
shift = 1/2
mycmap = copy.copy(plt.get_cmap("coolwarm")) #divergent colormap
fig = plt.figure(figsize=(5,5))
ax = fig.add_subplot(111,projection="3d")
phi = np.linspace(0,2*np.pi,res.shape[1])
theta = np.linspace(0,1*np.pi,res.shape[0])
n_phi = len(phi)
n_theta = len(theta)
# In order to draw gridlines, we add a colour below the current colormap
mycmap.set_under('k')
# draw a grid line every `ph_gs`/`th_gs` steps
ph_gs = int(2*np.pi/6/(phi[2]-phi[1]))+1 # 60 degree
th_gs = int(np.pi/6/(theta[2]-theta[1]))+1 # 30 degree
x = np.outer(np.cos(phi), np.sin(theta))
y = np.outer(np.sin(phi), np.sin(theta))
z = np.outer(np.ones(np.size(phi)), np.cos(theta))
ax.set_axis_off()
ax.view_init(0, 180) # theta, phi
inp = []
for j in range(0, n_phi):
for i in range(0, n_theta):
val = res[i,j] / scale + shift
# if theta and phi are on the gridlines, we set the value to -1
# such that we draw the dark color of the grid lines,
# but only if the value of the function is very small.
# Like that, the grid lines will be underneath the actual plot.
if ((j%ph_gs) < 0.01 and abs(val-shift) < 1E-1):
val = -1
if ((i%th_gs) < 0.01 and abs(val-shift) < 1E-1):
val = -1
inp.append([j, i, val])
inp = np.array(inp)
c = inp[:, 2].reshape((n_phi, n_theta))
ax.plot_surface(x, y, z, rstride=1, cstride=1,
cmap=mycmap, facecolors=mycmap(c), alpha=1, linewidth=0.1, antialiased=False, vmax=1, vmin=0
)
plt.savefig(Path(plot_path)/filename, dpi=200)
if plot_inline:
print(Path(plot_path)/filename)
plt.show()
plt.close(fig)
[25]:
plot_sphere("rf" / results_folder, "pictures", plot_ids=[0,9,10,19], plot_inline=False)
Video (2D)¶
In the end, we will create a video of the dynamics. For this we will reuse most of the functions from above. First of all, we increase the number of timesteps:
[26]:
qdyn_tlist = ( .24 , 240 ) #= ( T , Nt )
We also change the folder where to save the files and decrease the resolution a little (feel free play whith this)
[27]:
results_folder = Path("results_video")
N_theta = 300
N_phi = 150
… and finally set up the model and run the code:
[28]:
qdyn_model(
H, qdyn_tlist, psi0,
results_folder = results_folder,
print_time_needed = False,
picture_dims = (N_theta, N_phi),
exp_bloch = False
)
[29]:
!./prop_spin_sys
*** Read config file rf/config ***
*** Done reading config file ***
*** Initializing system ***
*** Initializing grid ***
No explicit grid found in config file! Initializing grid as one dimensional 1x1 fake grid
Initializing grid as 1D cartesian grid without mapping
Number of grid points was set to nr = 1 for dimension dim = 1! Initializing dimension 1 as a 1x1 fake-grid...
*** Initializing pulses ***
1 pulses in config file
Initializing pulse 1
setting pulse shape as constant 1.00 in [ 0.000E+00, 2.400E-01]
*** Initializing dynamical generator ***
*** Done with initialization ***
Initialising wigner symbols on 32 threads:
Calculating and writing the spin-wigner function took: 18.9903299808502
s
Calculating and writing the q-function took: 189.470355033875 s
Using the plot routine above, we plot the new results into another folder:
[30]:
%%capture
# `%%capture` suppresses the output and can be removed while playing around ;-)
plot_surface("rf" / results_folder, "pictures_video", plot_inline=False)
Compiling the video¶
Finally, we use ffmpeg to compile all plots into the videos qfct.mp4 and wigner.mp4, both located in the folder pictures_video.
[31]:
!ffmpeg -framerate 16 -i "pictures_video/qfct_%05d.png" -c:v libx264 -y pictures_video/qfct.mp4
ffmpeg version 3.4.6-0ubuntu0.18.04.1 Copyright (c) 2000-2019 the FFmpeg developers
built with gcc 7 (Ubuntu 7.3.0-16ubuntu3)
configuration: --prefix=/usr --extra-version=0ubuntu0.18.04.1 --toolchain=hardened --libdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --enable-gpl --disable-stripping --enable-avresample --enable-avisynth --enable-gnutls --enable-ladspa --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libcdio --enable-libflite --enable-libfontconfig --enable-libfreetype --enable-libfribidi --enable-libgme --enable-libgsm --enable-libmp3lame --enable-libmysofa --enable-libopenjpeg --enable-libopenmpt --enable-libopus --enable-libpulse --enable-librubberband --enable-librsvg --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libssh --enable-libtheora --enable-libtwolame --enable-libvorbis --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx265 --enable-libxml2 --enable-libxvid --enable-libzmq --enable-libzvbi --enable-omx --enable-openal --enable-opengl --enable-sdl2 --enable-libdc1394 --enable-libdrm --enable-libiec61883 --enable-chromaprint --enable-frei0r --enable-libopencv --enable-libx264 --enable-shared
libavutil 55. 78.100 / 55. 78.100
libavcodec 57.107.100 / 57.107.100
libavformat 57. 83.100 / 57. 83.100
libavdevice 57. 10.100 / 57. 10.100
libavfilter 6.107.100 / 6.107.100
libavresample 3. 7. 0 / 3. 7. 0
libswscale 4. 8.100 / 4. 8.100
libswresample 2. 9.100 / 2. 9.100
libpostproc 54. 7.100 / 54. 7.100
Input #0, image2, from 'pictures_video/qfct_%05d.png':
Duration: 00:00:15.00, start: 0.000000, bitrate: N/A
Stream #0:0: Video: png, rgba(pc), 1200x800 [SAR 7874:7874 DAR 3:2], 16 fps, 16 tbr, 16 tbn, 16 tbc
Stream mapping:
Stream #0:0 -> #0:0 (png (native) -> h264 (libx264))
Press [q] to stop, [?] for help
[libx264 @ 0x55992192ee00] using SAR=1/1
[libx264 @ 0x55992192ee00] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX
[libx264 @ 0x55992192ee00] profile High 4:4:4 Predictive, level 3.2, 4:4:4 8-bit
[libx264 @ 0x55992192ee00] 264 - core 152 r2854 e9a5903 - H.264/MPEG-4 AVC codec - Copyleft 2003-2017 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x1:0x111 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=0 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=4 threads=25 lookahead_threads=4 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=16 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
Output #0, mp4, to 'pictures_video/qfct.mp4':
Metadata:
encoder : Lavf57.83.100
Stream #0:0: Video: h264 (libx264) (avc1 / 0x31637661), yuv444p, 1200x800 [SAR 1:1 DAR 3:2], q=-1--1, 16 fps, 16384 tbn, 16 tbc
Metadata:
encoder : Lavc57.107.100 libx264
Side data:
cpb: bitrate max/min/avg: 0/0/0 buffer size: 0 vbv_delay: -1
frame= 240 fps=121 q=-1.0 Lsize= 178kB time=00:00:14.81 bitrate= 98.3kbits/s speed=7.49x
video:174kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 2.089669%
[libx264 @ 0x55992192ee00] frame I:1 Avg QP: 8.84 size: 5883
[libx264 @ 0x55992192ee00] frame P:60 Avg QP:12.97 size: 1361
[libx264 @ 0x55992192ee00] frame B:179 Avg QP:14.13 size: 503
[libx264 @ 0x55992192ee00] consecutive B-frames: 0.4% 0.0% 1.2% 98.3%
[libx264 @ 0x55992192ee00] mb I I16..4: 96.7% 0.0% 3.3%
[libx264 @ 0x55992192ee00] mb P I16..4: 5.5% 0.0% 0.1% P16..4: 8.9% 1.8% 0.2% 0.0% 0.0% skip:83.6%
[libx264 @ 0x55992192ee00] mb B I16..4: 0.3% 0.0% 0.0% B16..8: 8.9% 0.5% 0.0% direct: 0.0% skip:90.2% L0:42.6% L1:56.7% BI: 0.8%
[libx264 @ 0x55992192ee00] coded y,u,v intra: 0.8% 0.0% 0.1% inter: 0.0% 0.0% 0.0%
[libx264 @ 0x55992192ee00] i16 v,h,dc,p: 38% 46% 5% 11%
[libx264 @ 0x55992192ee00] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 30% 29% 32% 1% 2% 2% 2% 1% 1%
[libx264 @ 0x55992192ee00] Weighted P-Frames: Y:0.0% UV:0.0%
[libx264 @ 0x55992192ee00] ref P L0: 64.1% 1.7% 18.2% 16.0%
[libx264 @ 0x55992192ee00] ref B L0: 89.0% 9.6% 1.5%
[libx264 @ 0x55992192ee00] ref B L1: 95.8% 4.2%
[libx264 @ 0x55992192ee00] kb/s:94.68
[32]:
!ffmpeg -framerate 16 -i "pictures_video/wigner_%05d.png" -c:v libx264 -y pictures_video/wigner.mp4
ffmpeg version 3.4.6-0ubuntu0.18.04.1 Copyright (c) 2000-2019 the FFmpeg developers
built with gcc 7 (Ubuntu 7.3.0-16ubuntu3)
configuration: --prefix=/usr --extra-version=0ubuntu0.18.04.1 --toolchain=hardened --libdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --enable-gpl --disable-stripping --enable-avresample --enable-avisynth --enable-gnutls --enable-ladspa --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libcdio --enable-libflite --enable-libfontconfig --enable-libfreetype --enable-libfribidi --enable-libgme --enable-libgsm --enable-libmp3lame --enable-libmysofa --enable-libopenjpeg --enable-libopenmpt --enable-libopus --enable-libpulse --enable-librubberband --enable-librsvg --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libssh --enable-libtheora --enable-libtwolame --enable-libvorbis --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx265 --enable-libxml2 --enable-libxvid --enable-libzmq --enable-libzvbi --enable-omx --enable-openal --enable-opengl --enable-sdl2 --enable-libdc1394 --enable-libdrm --enable-libiec61883 --enable-chromaprint --enable-frei0r --enable-libopencv --enable-libx264 --enable-shared
libavutil 55. 78.100 / 55. 78.100
libavcodec 57.107.100 / 57.107.100
libavformat 57. 83.100 / 57. 83.100
libavdevice 57. 10.100 / 57. 10.100
libavfilter 6.107.100 / 6.107.100
libavresample 3. 7. 0 / 3. 7. 0
libswscale 4. 8.100 / 4. 8.100
libswresample 2. 9.100 / 2. 9.100
libpostproc 54. 7.100 / 54. 7.100
Input #0, image2, from 'pictures_video/wigner_%05d.png':
Duration: 00:00:15.00, start: 0.000000, bitrate: N/A
Stream #0:0: Video: png, rgba(pc), 1200x800 [SAR 7874:7874 DAR 3:2], 16 fps, 16 tbr, 16 tbn, 16 tbc
Stream mapping:
Stream #0:0 -> #0:0 (png (native) -> h264 (libx264))
Press [q] to stop, [?] for help
[libx264 @ 0x55a9600b5e00] using SAR=1/1
[libx264 @ 0x55a9600b5e00] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX
[libx264 @ 0x55a9600b5e00] profile High 4:4:4 Predictive, level 3.2, 4:4:4 8-bit
[libx264 @ 0x55a9600b5e00] 264 - core 152 r2854 e9a5903 - H.264/MPEG-4 AVC codec - Copyleft 2003-2017 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x1:0x111 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=0 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=4 threads=25 lookahead_threads=4 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=16 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
Output #0, mp4, to 'pictures_video/wigner.mp4':
Metadata:
encoder : Lavf57.83.100
Stream #0:0: Video: h264 (libx264) (avc1 / 0x31637661), yuv444p, 1200x800 [SAR 1:1 DAR 3:2], q=-1--1, 16 fps, 16384 tbn, 16 tbc
Metadata:
encoder : Lavc57.107.100 libx264
Side data:
cpb: bitrate max/min/avg: 0/0/0 buffer size: 0 vbv_delay: -1
frame= 240 fps=120 q=-1.0 Lsize= 275kB time=00:00:14.81 bitrate= 152.3kbits/s speed= 7.4x
video:272kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 1.338302%
[libx264 @ 0x55a9600b5e00] frame I:1 Avg QP: 8.50 size: 5648
[libx264 @ 0x55a9600b5e00] frame P:60 Avg QP:14.16 size: 3430
[libx264 @ 0x55a9600b5e00] frame B:179 Avg QP:13.05 size: 369
[libx264 @ 0x55a9600b5e00] consecutive B-frames: 0.4% 0.0% 1.2% 98.3%
[libx264 @ 0x55a9600b5e00] mb I I16..4: 96.4% 0.0% 3.6%
[libx264 @ 0x55a9600b5e00] mb P I16..4: 4.2% 0.0% 0.5% P16..4: 9.9% 4.5% 1.3% 0.0% 0.0% skip:79.6%
[libx264 @ 0x55a9600b5e00] mb B I16..4: 0.1% 0.0% 0.0% B16..8: 8.3% 0.2% 0.0% direct: 0.0% skip:91.4% L0:32.8% L1:66.1% BI: 1.1%
[libx264 @ 0x55a9600b5e00] coded y,u,v intra: 6.9% 2.1% 2.1% inter: 0.8% 0.2% 0.2%
[libx264 @ 0x55a9600b5e00] i16 v,h,dc,p: 52% 36% 5% 7%
[libx264 @ 0x55a9600b5e00] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 58% 14% 18% 2% 2% 2% 1% 2% 1%
[libx264 @ 0x55a9600b5e00] Weighted P-Frames: Y:0.0% UV:0.0%
[libx264 @ 0x55a9600b5e00] ref P L0: 63.9% 9.9% 16.0% 10.2%
[libx264 @ 0x55a9600b5e00] ref B L0: 89.1% 9.7% 1.2%
[libx264 @ 0x55a9600b5e00] ref B L1: 98.1% 1.9%
[libx264 @ 0x55a9600b5e00] kb/s:148.04