make[1]: Entering directory '$(@D)' Making check in config make[2]: Entering directory '$(@D)/config' make[2]: Nothing to be done for 'check'. make[2]: Leaving directory '$(@D)/config' Making check in m4 make[2]: Entering directory '$(@D)/m4' make[2]: Nothing to be done for 'check'. make[2]: Leaving directory '$(@D)/m4' Making check in term make[2]: Entering directory '$(@D)/term' make[2]: Nothing to be done for 'check'. make[2]: Leaving directory '$(@D)/term' Making check in src make[2]: Entering directory '$(@D)/src' Making timestamp.h fatal: not a git repository: '$(SOURCE_DIR)/.git' stat: cannot stat 'Makefile.am': No such file or directory /usr/gnu/bin/make check-recursive make[3]: Entering directory '$(@D)/src' Making check in wxterminal make[4]: Entering directory '$(@D)/src/wxterminal' make[4]: Nothing to be done for 'check'. make[4]: Leaving directory '$(@D)/src/wxterminal' Making check in qtterminal make[4]: Entering directory '$(@D)/src/qtterminal' make[4]: Nothing to be done for 'check'. make[4]: Leaving directory '$(@D)/src/qtterminal' make[4]: Entering directory '$(@D)/src' make[4]: Nothing to be done for 'check-am'. make[4]: Leaving directory '$(@D)/src' make[3]: Leaving directory '$(@D)/src' make[2]: Leaving directory '$(@D)/src' Making check in docs make[2]: Entering directory '$(@D)/docs' /usr/gnu/bin/make check-am make[3]: Entering directory '$(@D)/docs' /usr/gnu/bin/make check-local make[4]: Entering directory '$(@D)/docs' ./checkdoc < $(SOURCE_DIR)/docs/gnuplot.doc; \ if test $? -eq 0; then \ echo "PASS: gnuplot.doc"; \ else \ :; \ fi :313:D polargrid 4 :314:DB :315:D windrose 1 :316:D sectors 4 :317:DB :318:D sharpen 1 :319:D iris 2 :320:DB :321:D contourfill 4 :345:D convex_hull 2 :346:D mask_pm3d 3 :347:D smooth_path 2 :364:D named_palettes 4 :365:D viridis 1 :488:D watchpoints 2 :507:D epi_data 1 :4631:D watchpoints 2 spaces-only line :5065 :5597:D histogram_colors 1 :5681:D argb_hexdata 2 spaces-only line :5923 spaces-only line :6057 spaces-only line :6071 tab character in line :6526 tab character in line :6527 tab character in line :6528 tab character in line :6529 spaces-only line :6614 spaces-only line :7321 spaces-only line :7562 spaces-only line :7685 tab character in line :8094 :8583:D sharpen 1 :10952:D contours 5 :10953:D discrete 3 spaces-only line :10988 :11139:D contourfill 3 :11157:D dashtypes 2 :11469:D heatmap_points 1 :11470:D heatmap_points 2 :11471:D heatmap_points 3 :12326:D hidden 6 :14441:D viridis 1 spaces-only line :14796 :14913:D spotlight 1 :16595:D ttics 3 :18695:D viridis 1 PASS: gnuplot.doc make[4]: Leaving directory '$(@D)/docs' make[3]: Leaving directory '$(@D)/docs' make[2]: Leaving directory '$(@D)/docs' Making check in man make[2]: Entering directory '$(@D)/man' Making check in ja make[3]: Entering directory '$(@D)/man/ja' make[3]: Nothing to be done for 'check'. make[3]: Leaving directory '$(@D)/man/ja' make[3]: Entering directory '$(@D)/man' make[3]: Nothing to be done for 'check-am'. make[3]: Leaving directory '$(@D)/man' make[2]: Leaving directory '$(@D)/man' Making check in demo make[2]: Entering directory '$(@D)/demo' Making check in plugin make[3]: Entering directory '$(@D)/demo/plugin' make[3]: Nothing to be done for 'check'. make[3]: Leaving directory '$(@D)/demo/plugin' make[3]: Entering directory '$(@D)/demo' /usr/gnu/bin/make check-local make[4]: Entering directory '$(@D)/demo' ******************** file simple.dem ******************** QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-andreas' Hit return to continueHit return to continue"simple.dem" line 21: warning: Did you try to plot a complex-valued function? Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue******************** file controls.dem ******************** Hit return to continue******************** file electron.dem ******************** Hit return to continueHit return to continueHit return to continue******************** file using.dem ******************** Hit return to continueHit return to continue******************** file fillstyle.dem ******************** Now draw the boxes with solid fillNow draw the boxes with a black borderNow make the boxes a little less wideAnd now let's try a different fill densityNow draw the boxes with no borderOr maybe a pattern fill, instead?Finished this demo******************** file fillcvrs.dem ******************** Press Return to continuePress Return to continuePress Return to continuePress Return to continuePress Return to continuePress Return to continuePress Return to continueHit return to continueHit return to continueHit return to continueHit return to continue******************** file candlesticks.dem ******************** Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue******************** file autoscale.dem ******************** Hit return to continueHit return to continueHit return to continue******************** file bins.dem ************************* Hit return to continue******************** smooth splines ******************** various splines for smoothing Now apply a smoothing spline, weighted by 1/rel error (-> return)Make it smoother by changing the smoothing weights (-> return)Accentuate the relative changes with logscaling on yNow approximate the data with a bezier curve between the endpoints (-> return)You would rather use log-scales ? (-> return)Same thing in 3D - planar caseSame thing in 3D general caseHit return to continueHit to continueHit to continue to continue to continue****************** file convex_hull.dem ******************** to continue to continue to continue****************** file concave_hull.dem ******************** to continue to continue to continue to continue to continue****************** file mask_pm3d.dem ******************** # Curve 0 of 1, 9 points # Curve title: "Convex hull" # x y type -24.4571 1.79031 i -15.0611 21.314 i 7.8984 20.8154 i 23.4373 14.0071 i 19.34 -12.7014 i 12.2994 -22.9166 i 2.14596 -24.9946 i -5.11691 -18.2533 i -24.4571 1.79031 i to continue to continue to continue******************** file errorbars.dem ******************** various styles of errorbar Would you like boxes? (-> return)Only X-Bars? (-> return)Only Y-Bars? (-> return)Logscaled? (-> return)X as well? (-> return)If you like bars without tics (-> return)X-Bars only (-> return)Y-Bars only (-> return)filledcurve shaded error regionHit return to continue******************** file zerror.dem ******************** Hit return to continueHit return to continue******************** file fit.dem ******************** Some examples how data fitting using nonlinear least squares fit can be done. We fit a straight line to the data -- only as a demo without physical meaning. fit function: l(x) = y0 + m*x initial parameters: y0 = 1.1, m = -0.1 fit command: fit l(x) 'lcdemo.dat' via y0, m Now start fitting... (-> return)Press enter to proceed with the next example. Now use the real single-measurement weights from column 5. (Look at the file lcdemo.dat and compare the columns to see the difference.) Since these are weights we rescale the resulting parameter errors. fit settings: set fit errorscaling fit command : fit l(x) 'lcdemo.dat' using 1:2:5 yerr via y0, m Press enter to start the fit.Press enter to proceed with the next example. It's time now to try a more realistic model function: density(x) = x < Tc ? curve(x)+lowlin(x) : high(x) curve(x) = b*tanh(g*(Tc-x)) lowlin(x) = ml*(x-Tc) + dens_Tc high(x) = mh*(x-Tc) + dens_Tc density(x) is a function which shall fit the whole temperature range using a ?: expression. It contains 6 model parameters which will all be varied. Now take the start parameters out of the file 'start.par' and plot the function. fit command: fit density(x) 'lcdemo.dat' using 1:2:5 yerror via 'start.par' Press enter to start the fit.Press enter to proceed with the next example. Now a brief demonstration of 3d fitting. hemisphr.dat contains random points on a hemisphere of radius 1, but we let fit figure this out for us. It takes many iterations, so we limit them to 50. We also do not want intermediate results here. fit settings: set fit results maxiter 50 "fit.dem" line 112: warning: Did you try to plot a complex-valued function? fit function: h(x,y) = sqrt(r*r - (abs(x-x0))**2.2 - (abs(y-y0))**1.8) + z0 fit command : fit h(x,y) 'hemisphr.dat' using 1:2:3 via r, x0, y0, z0 Press enter to start the fit. After 50 iterations the fit converged. final sum of squares of residuals : 0.080165 rel. change during last iteration : 0 degrees of freedom (FIT_NDF) : 245 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 0.0180888 variance of residuals (reduced chisquare) = WSSR/ndf : 0.000327204 Final set of parameters Asymptotic Standard Error ======================= ========================== r = 1.00225 +/- 0.0003641 (0.03632%) x0 = -0.000450691 +/- 0.0003929 (87.17%) y0 = 0.000201292 +/- 0.0004735 (235.3%) z0 = -0.00104521 +/- 0.001526 (146%) correlation matrix of the fit parameters: r x0 y0 z0 r 1.000 x0 0.140 1.000 y0 0.085 -0.424 1.000 z0 -0.658 -0.045 -0.049 1.000 "fit.dem" line 120: warning: Did you try to plot a complex-valued function? Notice, however, that this would converge much faster when fitted in a more appropriate co-ordinate system: fit r 'hemisphr.dat' using 0:($1*$1+$2*$2+$3*$3) via r where we are fitting f(x)=r to the radii calculated as the data is read from the file. No x value is required in this case. (This is left as an exercise for the user). Another possibility is to prescale the variables (set fit prescale), which may improve convergence. fit settings: set fit maxiter 50 prescale fit command : fit h(x,y) 'hemisphr.dat' using 1:2:3 via r, x0,y0,z0 Press enter to proceed with the next example. Now an example on how to fit multi-branch functions. The model consists of two branches, the first describing longitudinal sound velocity as function of propagation direction (upper data, from dataset 1), the second describing transverse sound velocity (lower data, from dataset 0). The model uses these data in order to fit elastic stiffnesses which occur differently in both branches. fit function: f(x,y) = y==1 ? vlong(x) : vtrans(x) vlong(x) = sqrt(1.0/2.0/rho*1e9*(main(x) + mixed(x))) vtrans(x) = sqrt(1.0/2.0/rho*1e9*(main(x) - mixed(x))) y will be the index of the dataset. fit command: fit f(x,y) 'soundvel.dat' using 1:-2:2 via 'sound.par' Press enter to start the fit.iter chisq delta/lim lambda c33 c11 c44 c13 phi0 0 1.6651778833e+07 0.00e+00 1.06e+02 9.000000e+00 6.000000e+00 1.000000e+00 4.000000e+00 2.000000e+01 1 3.7115794520e+06 -3.49e+05 1.06e+01 1.107842e+01 5.715164e+00 1.112984e+00 5.269471e+00 5.489671e+00 2 3.0952217805e+05 -1.10e+06 1.06e+00 1.250349e+01 5.473118e+00 6.767568e-01 4.359096e+00 -2.308544e+00 3 7.9135498639e+04 -2.91e+05 1.06e-01 1.257557e+01 5.490760e+00 7.047546e-01 4.019414e+00 -3.385802e-01 4 7.8701397376e+04 -5.52e+02 1.06e-02 1.258878e+01 5.490036e+00 7.019290e-01 3.998785e+00 -3.997977e-01 5 7.8701391418e+04 -7.57e-03 1.06e-03 1.258874e+01 5.490047e+00 7.019482e-01 3.998746e+00 -3.995830e-01 iter chisq delta/lim lambda c33 c11 c44 c13 phi0 After 5 iterations the fit converged. final sum of squares of residuals : 78701.4 rel. change during last iteration : -7.57102e-08 degrees of freedom (FIT_NDF) : 144 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 23.3781 variance of residuals (reduced chisquare) = WSSR/ndf : 546.537 Final set of parameters Asymptotic Standard Error ======================= ========================== c33 = 12.5887 +/- 0.02898 (0.2302%) c11 = 5.49005 +/- 0.01846 (0.3363%) c44 = 0.701948 +/- 0.009755 (1.39%) c13 = 3.99875 +/- 0.03177 (0.7946%) phi0 = -0.399583 +/- 0.13 (32.54%) correlation matrix of the fit parameters: c33 c11 c44 c13 phi0 c33 1.000 c11 -0.066 1.000 c44 -0.198 -0.278 1.000 c13 -0.141 0.028 -0.086 1.000 phi0 0.114 -0.022 0.034 0.181 1.000 Look at the file 'hexa.fnc' to see how the branches are realized using the data index as input for a pseudo-3d fit. Press enter to proceed with the next example.Next we only use every fifth data point for fitting by using the 'every' keyword. Note the faster fit and its result. fit command: fit f(x,y) 'soundvel.dat' every 5 using 1:-2:2 via 'sound.par' Press enter to start the fit.iter chisq delta/lim lambda c33 c11 c44 c13 phi0 0 3.4156363488e+06 0.00e+00 1.06e+02 9.000000e+00 6.000000e+00 1.000000e+00 4.000000e+00 2.000000e+01 1 1.7633147044e+06 -9.37e+04 1.06e+01 1.068004e+01 5.714969e+00 1.220411e+00 5.416989e+00 1.563561e+01 2 3.6812403684e+05 -3.79e+05 1.06e+00 1.154575e+01 5.621298e+00 9.265286e-01 5.024562e+00 -3.686271e+00 3 2.6359224461e+04 -1.30e+06 1.06e-01 1.253003e+01 5.480326e+00 6.995740e-01 4.092691e+00 -5.722734e-02 4 1.9074727803e+04 -3.82e+04 1.06e-02 1.254656e+01 5.491040e+00 7.055514e-01 3.941309e+00 -9.110937e-01 5 1.9071441847e+04 -1.72e+01 1.06e-03 1.254887e+01 5.490704e+00 7.054929e-01 3.937661e+00 -8.989885e-01 6 1.9071441717e+04 -6.83e-04 1.06e-04 1.254886e+01 5.490728e+00 7.054880e-01 3.937655e+00 -8.989317e-01 iter chisq delta/lim lambda c33 c11 c44 c13 phi0 After 6 iterations the fit converged. final sum of squares of residuals : 19071.4 rel. change during last iteration : -6.82518e-09 degrees of freedom (FIT_NDF) : 26 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 27.0835 variance of residuals (reduced chisquare) = WSSR/ndf : 733.517 Final set of parameters Asymptotic Standard Error ======================= ========================== c33 = 12.5489 +/- 0.07395 (0.5893%) c11 = 5.49073 +/- 0.04794 (0.8732%) c44 = 0.705488 +/- 0.02385 (3.381%) c13 = 3.93766 +/- 0.08229 (2.09%) phi0 = -0.898932 +/- 0.3067 (34.11%) correlation matrix of the fit parameters: c33 c11 c44 c13 phi0 c33 1.000 c11 -0.067 1.000 c44 -0.227 -0.251 1.000 c13 -0.196 0.051 -0.066 1.000 phi0 0.086 0.006 -0.005 0.147 1.000 When you compare the results (see 'fit.log') you will note that the error of the fitted parameters have become larger, and the quality of the plot is only slightly affected. Press enter to proceed with the next example. By marking some parameters as '# FIXED' in the parameter file, you fit only the others (c44 and c13 are fixed here). Press enter to start the fit.iter chisq delta/lim lambda c33 c11 phi0 0 9.7945909430e+06 0.00e+00 7.60e+01 9.000000e+00 6.000000e+00 1.000000e-04 1 5.6703596465e+05 -1.63e+06 7.60e+00 1.220149e+01 5.310817e+00 -7.224014e-01 2 5.3024065948e+05 -6.94e+03 7.60e-01 1.240113e+01 5.340579e+00 -1.080402e-01 3 5.3014685038e+05 -1.77e+01 7.60e-02 1.240095e+01 5.340080e+00 -1.667665e-01 4 5.3014624975e+05 -1.13e-01 7.60e-03 1.240106e+01 5.340147e+00 -1.620752e-01 iter chisq delta/lim lambda c33 c11 phi0 After 4 iterations the fit converged. final sum of squares of residuals : 530146 rel. change during last iteration : -1.13295e-06 degrees of freedom (FIT_NDF) : 146 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 60.2589 variance of residuals (reduced chisquare) = WSSR/ndf : 3631.14 Final set of parameters Asymptotic Standard Error ======================= ========================== c33 = 12.4011 +/- 0.07251 (0.5847%) c11 = 5.34015 +/- 0.0462 (0.8651%) phi0 = -0.162075 +/- 0.3543 (218.6%) correlation matrix of the fit parameters: c33 c11 phi0 c33 1.000 c11 -0.135 1.000 phi0 0.153 0.001 1.000 This has the same effect as specifying only the real free parameters by the 'via' syntax: fit f(x) 'soundvel.dat' via c33, c11, phi0 Press enter to proceed with the next example. Here comes an example of a rather complex function. First a plot with all parameters set to initial values. Now fit the model function to the data. fit settings: set fit limit 1e-10 fit function: R(x) = sinh(A*a(x)) * exp(-1.*A*(1.+a(x))) a(x) = W(x) * Q(tc) / mu W(x) = 1./(sqrt(2.*pi)*eta) * exp( -1. * x**2 / (2.*eta**2) ) initial parameters: eta = 0.00012 tc = 0.0018 fit command : fit R(x) 'moli3.dat' u 1:2:3 zerror via eta, tc now start fitting... (-> return)iter chisq delta/lim lambda eta tc 0 1.1441984213e+04 0.00e+00 2.76e+05 1.200000e-04 1.800000e-03 1 5.3171854336e+03 -1.15e+10 2.76e+04 1.043852e-04 1.837367e-03 2 4.6879093617e+03 -1.34e+09 2.76e+03 1.018093e-04 2.009651e-03 3 4.6734120845e+03 -3.10e+07 2.76e+02 1.010031e-04 2.024420e-03 4 4.6729937953e+03 -8.95e+05 2.76e+01 1.008229e-04 2.021774e-03 5 4.6729736309e+03 -4.32e+04 2.76e+00 1.007894e-04 2.021375e-03 6 4.6729718879e+03 -3.73e+03 2.76e-01 1.007831e-04 2.021299e-03 7 4.6729716327e+03 -5.46e+02 2.76e-02 1.007819e-04 2.021285e-03 8 4.6729715874e+03 -9.70e+01 2.76e-03 1.007817e-04 2.021282e-03 9 4.6729715790e+03 -1.80e+01 2.76e-04 1.007817e-04 2.021282e-03 10 4.6729715774e+03 -3.37e+00 2.76e-05 1.007817e-04 2.021282e-03 11 4.6729715771e+03 -6.32e-01 2.76e-06 1.007817e-04 2.021282e-03 iter chisq delta/lim lambda eta tc After 11 iterations the fit converged. final sum of squares of residuals : 4672.97 rel. change during last iteration : -6.32403e-11 degrees of freedom (FIT_NDF) : 123 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 6.16374 variance of residuals (reduced chisquare) = WSSR/ndf : 37.9916 p-value of the Chisq distribution (FIT_P) : 0 Final set of parameters Standard Deviation ======================= ========================== eta = 0.000100782 +/- 3.184e-07 (0.3159%) tc = 0.00202128 +/- 1.282e-05 (0.6342%) correlation matrix of the fit parameters: eta tc eta 1.000 tc 0.426 1.000 Looking at the plot of the resulting fit curve, you can see that this function doesn't really fit this set of data points. This would normally be a reason to to check for measurement problems not yet accounted for, and maybe even re-think the theoretic prediction in use. Press enter to proceed with the next example. Next we show a fit with three independent variables. The file fit3.dat has four columns, with values of the three independent variables x, y and t, and the'resulting value z. The data lines are in four sections, with t being constant within each section. The sections are separated by two blank lines, so we can select sections with "index" modifiers. Here are the data in the first section, where t = -3. We will fit the function a0/(1 + a1*x**2 + a2*y**2) to these data. Since at this point we have two independent variables, our "using" spec has four entries, representing x:y:z:s (where s is the estimated error in the z value). fit function: f1(x,y)=a0/(1+a1*x**2+a2*y**2) fit command : fit f1(x,y) 'fit3.dat' index 0 using 1:2:4 via a0,a1,a2 Press enter to start the fit.iter chisq delta/lim lambda a0 a1 a2 0 1.9200759829e+02 0.00e+00 1.08e+00 1.000000e+00 1.000000e-01 1.000000e-01 * 2.1341288746e+05 9.99e+09 1.08e+01 -1.996446e+00 -9.317214e-02 -6.894074e-02 1 1.2155812773e+02 -5.80e+09 1.08e+00 6.747760e-01 3.330668e-01 3.459786e-01 2 6.4591509465e+00 -1.78e+11 1.08e-01 -2.166519e+00 4.014935e-01 5.408270e-01 3 1.0813895568e+00 -4.97e+10 1.08e-02 -3.016252e+00 5.534097e-01 4.635940e-01 4 1.0604896443e+00 -1.97e+08 1.08e-03 -3.021526e+00 5.281087e-01 4.842650e-01 5 1.0604647203e+00 -2.35e+05 1.08e-04 -3.022777e+00 5.291208e-01 4.850168e-01 6 1.0604647180e+00 -2.20e+01 1.08e-05 -3.022759e+00 5.291036e-01 4.850186e-01 7 1.0604647180e+00 -2.62e-01 1.08e-06 -3.022760e+00 5.291039e-01 4.850187e-01 iter chisq delta/lim lambda a0 a1 a2 After 7 iterations the fit converged. final sum of squares of residuals : 1.06046 rel. change during last iteration : -2.61567e-11 degrees of freedom (FIT_NDF) : 118 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 0.0947997 variance of residuals (reduced chisquare) = WSSR/ndf : 0.00898699 Final set of parameters Asymptotic Standard Error ======================= ========================== a0 = -3.02276 +/- 0.05612 (1.857%) a1 = 0.529104 +/- 0.02538 (4.798%) a2 = 0.485019 +/- 0.02338 (4.821%) correlation matrix of the fit parameters: a0 a1 a2 a0 1.000 a1 -0.636 1.000 a2 -0.638 0.235 1.000 Press enter to proceed with the next example. Here is the last set of data where t = 3. We fit the same function to this set. fit function: f1(x,y)=a0/(1+a1*x**2+a2*y**2) fit command : fit f1(x,y) 'fit3.dat' index 3 using 1:2:4 via a0,a1,a2 Press enter to start the fit.iter chisq delta/lim lambda a0 a1 a2 0 2.7120346202e+02 0.00e+00 4.18e-01 -3.022760e+00 5.291039e-01 4.850187e-01 1 5.0488012023e+00 -5.27e+11 4.18e-02 2.596854e+00 6.962571e-01 5.636897e-01 2 1.4652164280e+00 -2.45e+10 4.18e-03 3.093112e+00 4.150048e-01 5.365566e-01 3 1.1757252546e+00 -2.46e+09 4.18e-04 3.115259e+00 4.994275e-01 5.471436e-01 4 1.1726111180e+00 -2.66e+07 4.18e-05 3.120030e+00 5.112617e-01 5.467837e-01 5 1.1726107387e+00 -3.23e+03 4.18e-06 3.119771e+00 5.112950e-01 5.466635e-01 6 1.1726107381e+00 -5.57e+00 4.18e-07 3.119777e+00 5.112965e-01 5.466665e-01 * 1.1726107381e+00 1.67e-01 4.18e-06 3.119777e+00 5.112964e-01 5.466664e-01 * 1.1726107381e+00 1.67e-01 4.18e-05 3.119777e+00 5.112964e-01 5.466664e-01 * 1.1726107381e+00 1.67e-01 4.18e-04 3.119777e+00 5.112964e-01 5.466664e-01 * 1.1726107381e+00 1.67e-01 4.18e-03 3.119777e+00 5.112964e-01 5.466664e-01 * 1.1726107381e+00 1.67e-01 4.18e-02 3.119777e+00 5.112964e-01 5.466664e-01 * 1.1726107381e+00 1.67e-01 4.18e-01 3.119777e+00 5.112964e-01 5.466664e-01 * 1.1726107381e+00 1.54e-01 4.18e+00 3.119777e+00 5.112964e-01 5.466664e-01 * 1.1726107381e+00 3.65e-02 4.18e+01 3.119777e+00 5.112964e-01 5.466665e-01 * 1.1726107381e+00 8.58e-04 4.18e+02 3.119777e+00 5.112965e-01 5.466665e-01 * 1.1726107381e+00 7.57e-06 4.18e+03 3.119777e+00 5.112965e-01 5.466665e-01 7 1.1726107381e+00 -1.89e-06 4.18e+02 3.119777e+00 5.112965e-01 5.466665e-01 iter chisq delta/lim lambda a0 a1 a2 After 7 iterations the fit converged. final sum of squares of residuals : 1.17261 rel. change during last iteration : -1.89359e-16 degrees of freedom (FIT_NDF) : 117 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 0.100112 variance of residuals (reduced chisquare) = WSSR/ndf : 0.0100223 Final set of parameters Asymptotic Standard Error ======================= ========================== a0 = 3.11978 +/- 0.0752 (2.41%) a1 = 0.511296 +/- 0.02835 (5.544%) a2 = 0.546667 +/- 0.03019 (5.522%) correlation matrix of the fit parameters: a0 a1 a2 a0 1.000 a1 0.718 1.000 a2 0.716 0.376 1.000 Press enter to proceed with the next example. We also have data for several intermediate values of t. We will fit the function f(x,y,t)=a0*t/(1+a1*x**2+a2*y**2) to all the data. fit function: f(x,y,t)=a0*t/(1+a1*x**2+a2*y**2) fit command : fit f(x,y,t) 'fit3.dat' u 1:2:3:4 via a0,a1,a2 Press enter to start the fit.iter chisq delta/lim lambda a0 a1 a2 0 6.6327563650e+02 0.00e+00 9.11e-01 3.119777e+00 5.112965e-01 5.466665e-01 1 4.6639172676e+00 -1.41e+12 9.11e-02 1.058306e+00 5.232988e-01 5.443153e-01 2 4.5674642800e+00 -2.11e+08 9.11e-03 1.021357e+00 5.174850e-01 5.083111e-01 3 4.5674523432e+00 -2.61e+04 9.11e-04 1.021796e+00 5.177911e-01 5.088947e-01 4 4.5674523419e+00 -2.84e+00 9.11e-05 1.021790e+00 5.177820e-01 5.088906e-01 5 4.5674523419e+00 -4.67e-02 9.11e-06 1.021790e+00 5.177821e-01 5.088907e-01 iter chisq delta/lim lambda a0 a1 a2 After 5 iterations the fit converged. final sum of squares of residuals : 4.56745 rel. change during last iteration : -4.67264e-12 degrees of freedom (FIT_NDF) : 480 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 0.0975476 variance of residuals (reduced chisquare) = WSSR/ndf : 0.00951553 Final set of parameters Asymptotic Standard Error ======================= ========================== a0 = 1.02179 +/- 0.01414 (1.384%) a1 = 0.517782 +/- 0.01766 (3.41%) a2 = 0.508891 +/- 0.01737 (3.414%) correlation matrix of the fit parameters: a0 a1 a2 a0 1.000 a1 0.669 1.000 a2 0.669 0.289 1.000 Here are all the data together. You can use ranges to rename variables and/or limit the data included in the fit. The first range corresponds to the first "using" entry, etc. For example, we could have gotten the same fit like this: fit [lon=*:*][lat=*:*][time=*:*] \ a0*time/(1 + a1*lon**2 + a2*lat**2) \ "fit3.dat" u 1:2:3:4 via a0,a1,a2 Press enter to proceed with the next example. The fit command can handle errors in the independent variable, too. The problem shown here is Pearson's data with York's weights. First draw the data with uncertainties and the initial function. Press enter to fit the data using no error values Press enter to fit the data using no error valueslambda start value set: 1 iter chisq delta/lim lambda a1 a2 0 4.6100000000e+00 0.00e+00 1.00e+00 5.000000e+00 -5.000000e-01 1 8.0189138044e-01 -4.75e+08 1.00e-01 5.741046e+00 -5.350813e-01 2 8.0066352303e-01 -1.53e+05 1.00e-02 5.761170e+00 -5.395735e-01 3 8.0066352224e-01 -9.89e-02 1.00e-03 5.761185e+00 -5.395773e-01 iter chisq delta/lim lambda a1 a2 After 3 iterations the fit converged. final sum of squares of residuals : 0.800664 rel. change during last iteration : -9.88845e-10 degrees of freedom (FIT_NDF) : 8 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 0.316359 variance of residuals (reduced chisquare) = WSSR/ndf : 0.100083 Final set of parameters Asymptotic Standard Error ======================= ========================== a1 = 5.76119 +/- 0.1895 (3.289%) a2 = -0.539577 +/- 0.04213 (7.807%) correlation matrix of the fit parameters: a1 a2 a1 1.000 a2 -0.849 1.000 Press enter to fit the data using only the uncertainties of the y-values. Press enter to fit the data using only the uncertainties of the y-values.lambda start value set: 1 iter chisq delta/lim lambda a1y a2y 0 1.4962550000e+02 0.00e+00 1.00e+00 5.000000e+00 -5.000000e-01 1 3.4345697872e+01 -3.36e+08 1.00e-01 6.095602e+00 -6.101483e-01 2 3.4345207498e+01 -1.43e+03 1.00e-02 6.100109e+00 -6.108129e-01 3 3.4345207498e+01 -4.10e-06 1.00e-03 6.100109e+00 -6.108130e-01 iter chisq delta/lim lambda a1y a2y After 3 iterations the fit converged. final sum of squares of residuals : 34.3452 rel. change during last iteration : -4.09628e-14 degrees of freedom (FIT_NDF) : 8 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 2.07199 variance of residuals (reduced chisquare) = WSSR/ndf : 4.29315 p-value of the Chisq distribution (FIT_P) : 3.51726e-05 Final set of parameters Standard Deviation ======================= ========================== a1y = 6.10011 +/- 0.2047 (3.355%) a2y = -0.610813 +/- 0.03009 (4.926%) correlation matrix of the fit parameters: a1y a2y a1y 1.000 a2y -0.985 1.000 Press enter to fit the data using the uncertainties of the x and y values. Press enter to fit the data using the uncertainties of the x and y values.lambda start value set: 1 iter chisq delta/lim lambda a1xy a2xy 0 6.3259837889e+01 0.00e+00 1.00e+00 5.000000e+00 -5.000000e-01 1 1.1957122097e+01 -4.29e+08 1.00e-01 5.395533e+00 -4.633919e-01 2 1.1956475893e+01 -5.40e+03 1.00e-02 5.396066e+00 -4.634515e-01 * 1.1956484542e+01 7.23e+01 1.00e-01 5.396062e+00 -4.634507e-01 * 1.1956484540e+01 7.23e+01 1.00e+00 5.396062e+00 -4.634507e-01 * 1.1956484400e+01 7.12e+01 1.00e+01 5.396062e+00 -4.634507e-01 * 1.1956479134e+01 2.71e+01 1.00e+02 5.396064e+00 -4.634512e-01 * 1.1956475943e+01 4.19e-01 1.00e+03 5.396066e+00 -4.634515e-01 * 1.1956475893e+01 4.16e-03 1.00e+04 5.396066e+00 -4.634515e-01 * 1.1956475893e+01 4.03e-05 1.00e+05 5.396066e+00 -4.634515e-01 3 1.1956475893e+01 -1.63e-07 1.00e+04 5.396066e+00 -4.634515e-01 iter chisq delta/lim lambda a1xy a2xy After 3 iterations the fit converged. final sum of squares of residuals : 11.9565 rel. change during last iteration : -1.63425e-15 degrees of freedom (FIT_NDF) : 8 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 1.22252 variance of residuals (reduced chisquare) = WSSR/ndf : 1.49456 p-value of the Chisq distribution (FIT_P) : 0.153156 Final set of parameters Standard Deviation ======================= ========================== a1xy = 5.39607 +/- 0.2957 (5.479%) a2xy = -0.463451 +/- 0.0578 (12.47%) correlation matrix of the fit parameters: a1xy a2xy a1xy 1.000 a2xy -0.964 1.000 Summary of the fit results: a1 a2 a1_err a2_err ------------------------------------------------------------------------ initial values 5.000e+00 -5.00e-01 our result 5.396e+00 -4.63e-01 2.957e-01 5.78e-02 ROOT Minuit 5.480e+00 -4.81e-01 2.926e-01 5.76e-02 ------------------------------------------------------------------------ You can have a look at all previous fit results by looking into the file 'fit.log' (or whatever you defined the environment variable 'FIT_LOG' to). Remember that this file will always be appended to, so remove it from time to time. Done with fitting demo (-> return)"fitmulti.dem" line 83: warning: > Implied independent variable y not found in fit function. > Assuming version 4 syntax with zerror in column 3 but no zerror keyword. iter chisq delta/lim lambda c1 0 2.1892940362e+01 0.00e+00 6.81e-01 1.000000e+00 1 4.5233347856e-02 -4.83e+07 6.81e-02 2.431818e+00 2 1.0247233379e-08 -4.41e+11 6.81e-03 2.499968e+00 3 2.3236116830e-19 -4.41e+15 6.81e-04 2.500000e+00 4 1.9793013150e-28 -1.17e+14 6.81e-05 2.500000e+00 5 1.9793013150e-28 0.00e+00 6.81e-06 2.500000e+00 iter chisq delta/lim lambda c1 After 5 iterations the fit converged. final sum of squares of residuals : 1.9793e-28 rel. change during last iteration : 0 degrees of freedom (FIT_NDF) : 20 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 3.14587e-15 variance of residuals (reduced chisquare) = WSSR/ndf : 9.89651e-30 p-value of the Chisq distribution (FIT_P) : 1 Final set of parameters Standard Deviation ======================= ========================== c1 = 2.5 +/- 0.3206 (12.82%) --------------------------------------------------------- 1D fit: expected 2.5 c1 = 2.5 --------------------------------------------------------- Hit return to try for a 2D fit"fitmulti.dem" line 99: warning: > Implied independent variable t not found in fit function. > Assuming version 4 syntax with zerror in column 4 but no zerror keyword. iter chisq delta/lim lambda c1 c2 0 1.8402267568e+01 0.00e+00 5.08e-01 1.000000e+00 1.000000e+00 1 5.1495377021e-01 -3.47e+06 5.08e-02 2.363854e+00 -7.849976e-01 2 1.4278686826e-05 -3.61e+09 5.08e-03 2.499431e+00 -1.794590e+00 3 4.0743857030e-14 -3.50e+13 5.08e-04 2.500000e+00 -1.800000e+00 4 1.1747755312e-26 -3.47e+17 5.08e-05 2.500000e+00 -1.800000e+00 5 1.3330381888e-28 -8.71e+06 5.08e-06 2.500000e+00 -1.800000e+00 6 1.2866926106e-28 -3.60e+03 5.08e-07 2.500000e+00 -1.800000e+00 7 1.2866926106e-28 0.00e+00 5.08e-08 2.500000e+00 -1.800000e+00 iter chisq delta/lim lambda c1 c2 After 7 iterations the fit converged. final sum of squares of residuals : 1.28669e-28 rel. change during last iteration : 0 degrees of freedom (FIT_NDF) : 18 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 2.67363e-15 variance of residuals (reduced chisquare) = WSSR/ndf : 7.14829e-30 p-value of the Chisq distribution (FIT_P) : 1 Final set of parameters Standard Deviation ======================= ========================== c1 = 2.5 +/- 0.3503 (14.01%) c2 = -1.8 +/- 1.432 (79.57%) correlation matrix of the fit parameters: c1 c2 c1 1.000 c2 -0.403 1.000 --------------------------------------------------------- 2D fit: expected 2.5 c1 = 2.5 2D fit: expected -1.8 c2 = -1.80000000000001 --------------------------------------------------------- Hit return to try for a 3D fit"fitmulti.dem" line 115: warning: > Implied independent variable x3 not found in fit function. > Assuming version 4 syntax with zerror in column 5 but no zerror keyword. iter chisq delta/lim lambda c1 c2 c3 0 5.0042107018e+05 0.00e+00 1.36e+00 1.000000e+00 1.000000e+00 1.000000e+00 1 1.5648231886e+02 -3.20e+08 1.36e-01 2.667794e+00 2.485896e-01 6.876787e+01 2 9.5026215217e-04 -1.65e+10 1.36e-02 2.498116e+00 -1.774715e+00 6.999949e+01 3 1.4986897994e-11 -6.34e+12 1.36e-03 2.500000e+00 -1.799997e+00 7.000000e+01 4 2.5594986155e-23 -5.86e+16 1.36e-04 2.500000e+00 -1.800000e+00 7.000000e+01 5 1.5865767741e-24 -1.51e+06 1.36e-05 2.500000e+00 -1.800000e+00 7.000000e+01 6 1.5657265887e-24 -1.33e+03 1.36e-06 2.500000e+00 -1.800000e+00 7.000000e+01 * 1.5852570098e-24 1.23e+03 1.36e-05 2.500000e+00 -1.800000e+00 7.000000e+01 * 1.5852570098e-24 1.23e+03 1.36e-04 2.500000e+00 -1.800000e+00 7.000000e+01 * 1.5852570098e-24 1.23e+03 1.36e-03 2.500000e+00 -1.800000e+00 7.000000e+01 * 1.5852570098e-24 1.23e+03 1.36e-02 2.500000e+00 -1.800000e+00 7.000000e+01 * 1.5852570098e-24 1.23e+03 1.36e-01 2.500000e+00 -1.800000e+00 7.000000e+01 * 1.5852570098e-24 1.23e+03 1.36e+00 2.500000e+00 -1.800000e+00 7.000000e+01 * 1.5854870121e-24 1.25e+03 1.36e+01 2.500000e+00 -1.800000e+00 7.000000e+01 7 1.5657265887e-24 0.00e+00 1.36e+00 2.500000e+00 -1.800000e+00 7.000000e+01 iter chisq delta/lim lambda c1 c2 c3 After 7 iterations the fit converged. final sum of squares of residuals : 1.56573e-24 rel. change during last iteration : 0 degrees of freedom (FIT_NDF) : 18 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 2.94932e-13 variance of residuals (reduced chisquare) = WSSR/ndf : 8.69848e-26 p-value of the Chisq distribution (FIT_P) : 1 Final set of parameters Standard Deviation ======================= ========================== c1 = 2.5 +/- 0.3341 (13.36%) c2 = -1.8 +/- 0.8232 (45.73%) c3 = 70 +/- 0.09942 (0.142%) correlation matrix of the fit parameters: c1 c2 c3 c1 1.000 c2 -0.228 1.000 c3 -0.144 -0.089 1.000 --------------------------------------------------------- 3D fit: expected 2.5 c1 = 2.50000000000001 3D fit: expected -1.8 c2 = -1.80000000000011 3D fit: expected 70.0 c3 = 70.0 --------------------------------------------------------- Hit return to try for a 4D fit"fitmulti.dem" line 131: warning: > Implied independent variable x4 not found in fit function. > Assuming version 4 syntax with zerror in column 6 but no zerror keyword. iter chisq delta/lim lambda c1 c2 c3 c4 0 5.0434414591e+05 0.00e+00 1.23e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1 1.0737393796e+02 -4.70e+08 1.23e-01 2.637883e+00 2.081428e-01 6.896240e+01 -3.465848e+00 2 7.6766393648e-04 -1.40e+10 1.23e-02 2.498380e+00 -1.775303e+00 6.999930e+01 -3.204055e+00 3 1.1976288234e-11 -6.41e+12 1.23e-03 2.500000e+00 -1.799997e+00 7.000000e+01 -3.200001e+00 4 1.9635651177e-23 -6.10e+16 1.23e-04 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 5 6.5436327632e-25 -2.90e+06 1.23e-05 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 6 6.5213001110e-25 -3.42e+02 1.23e-06 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 7 6.5184207687e-25 -4.42e+01 1.23e-07 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 8 6.4608339226e-25 -8.91e+02 1.23e-08 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 * 6.5173794723e-25 8.68e+02 1.23e-07 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 * 6.5173794723e-25 8.68e+02 1.23e-06 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 * 6.5173794723e-25 8.68e+02 1.23e-05 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 * 6.5173794723e-25 8.68e+02 1.23e-04 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 * 6.5173794723e-25 8.68e+02 1.23e-03 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 * 6.5173794723e-25 8.68e+02 1.23e-02 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 * 6.5173794723e-25 8.68e+02 1.23e-01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 * 6.5173794723e-25 8.68e+02 1.23e+00 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 * 6.5178922319e-25 8.75e+02 1.23e+01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 9 6.4608339226e-25 0.00e+00 1.23e+00 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 iter chisq delta/lim lambda c1 c2 c3 c4 After 9 iterations the fit converged. final sum of squares of residuals : 6.46083e-25 rel. change during last iteration : 0 degrees of freedom (FIT_NDF) : 17 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 1.94948e-13 variance of residuals (reduced chisquare) = WSSR/ndf : 3.80049e-26 p-value of the Chisq distribution (FIT_P) : 1 Final set of parameters Standard Deviation ======================= ========================== c1 = 2.5 +/- 0.3341 (13.36%) c2 = -1.8 +/- 0.8947 (49.71%) c3 = 70 +/- 0.1026 (0.1466%) c4 = -3.2 +/- 0.3306 (10.33%) correlation matrix of the fit parameters: c1 c2 c3 c4 c1 1.000 c2 -0.214 1.000 c3 -0.137 -0.176 1.000 c4 0.010 -0.392 0.247 1.000 --------------------------------------------------------- 4D fit: expected 2.5 c1 = 2.4999999999999 4D fit: expected -1.8 c2 = -1.80000000000033 4D fit: expected 70.0 c3 = 70.0 4D fit: expected -3.2 c4 = -3.1999999999997 --------------------------------------------------------- Hit return to try for a 5D fit"fitmulti.dem" line 149: warning: > Implied independent variable x5 not found in fit function. > Assuming version 4 syntax with zerror in column 7 but no zerror keyword. iter chisq delta/lim lambda c1 c2 c3 c4 c5 0 3.5652827793e+05 0.00e+00 2.08e+01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1 8.2628175865e+04 -3.31e+05 2.08e+00 1.131551e+00 1.200143e+00 4.015282e+00 1.106540e+00 3.274197e+00 2 2.8797393982e+03 -2.77e+06 2.08e-01 2.268438e+00 3.004089e+00 5.746308e+01 -1.174504e+00 9.548228e-01 3 5.7761101766e-02 -4.99e+09 2.08e-02 2.484844e+00 -1.615778e+00 6.995937e+01 -3.211773e+00 4.017438e-01 4 5.5575896136e-09 -1.04e+12 2.08e-03 2.499994e+00 -1.799932e+00 6.999999e+01 -3.200009e+00 4.000002e-01 5 7.5296124774e-20 -7.38e+15 2.08e-04 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 6 6.8714202466e-25 -1.10e+10 2.08e-05 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 7 6.6497897753e-25 -3.33e+03 2.08e-06 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 * 6.6506654109e-25 1.32e+01 2.08e-05 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 * 6.6506654109e-25 1.32e+01 2.08e-04 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 * 6.6506654109e-25 1.32e+01 2.08e-03 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 * 6.6506654109e-25 1.32e+01 2.08e-02 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 * 6.6506654109e-25 1.32e+01 2.08e-01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 * 6.6500422107e-25 3.80e+00 2.08e+00 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 8 6.4747494290e-25 -2.70e+03 2.08e-01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 * 6.9279657963e-25 6.54e+03 2.08e+00 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 * 6.6666004012e-25 2.88e+03 2.08e+01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 * 6.6524640137e-25 2.67e+03 2.08e+02 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 * 6.5555287857e-25 1.23e+03 2.08e+03 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 9 6.4747494290e-25 0.00e+00 2.08e+02 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 iter chisq delta/lim lambda c1 c2 c3 c4 c5 After 9 iterations the fit converged. final sum of squares of residuals : 6.47475e-25 rel. change during last iteration : 0 degrees of freedom (FIT_NDF) : 16 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 2.01165e-13 variance of residuals (reduced chisquare) = WSSR/ndf : 4.04672e-26 p-value of the Chisq distribution (FIT_P) : 1 Final set of parameters Standard Deviation ======================= ========================== c1 = 2.5 +/- 0.3351 (13.41%) c2 = -1.8 +/- 0.9103 (50.57%) c3 = 70 +/- 0.2366 (0.338%) c4 = -3.2 +/- 0.3331 (10.41%) c5 = 0.4 +/- 0.01144 (2.86%) correlation matrix of the fit parameters: c1 c2 c3 c4 c5 c1 1.000 c2 -0.224 1.000 c3 0.011 -0.241 1.000 c4 0.001 -0.360 -0.004 1.000 c5 -0.078 0.184 -0.901 0.123 1.000 --------------------------------------------------------- 5D fit: expected 2.5 c1 = 2.4999999999999 5D fit: expected -1.8 c2 = -1.80000000000034 5D fit: expected 70.0 c3 = 70.0 5D fit: expected -3.2 c4 = -3.1999999999997 5D fit: expected 0.4 c5 = 0.4 --------------------------------------------------------- Hit return to try for a 6D fitThis 6D fit will fail in version 4 but version 5 can handle more parameters "fitmulti.dem" line 171: warning: > Implied independent variable not found in fit function. > Assuming version 4 syntax with zerror in column 8 but no zerror keyword. iter chisq delta/lim lambda c1 c2 c3 c4 c5 c6 0 3.4961271093e+05 0.00e+00 1.94e+01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1 7.6637354181e+04 -3.56e+05 1.94e+00 9.608331e-01 1.184762e+00 4.200546e+00 1.106839e+00 3.265869e+00 9.641673e-01 2 2.4111848689e+03 -3.08e+06 1.94e-01 1.293024e+00 2.647546e+00 5.815961e+01 -1.307965e+00 9.256806e-01 1.373154e-01 3 4.0425600039e-02 -5.96e+09 1.94e-02 2.493419e+00 -1.646308e+00 6.996638e+01 -3.211246e+00 4.014427e-01 -1.585899e-01 4 7.1103832630e-08 -5.69e+10 1.94e-03 2.500262e+00 -1.799903e+00 7.000002e+01 -3.200052e+00 3.999987e-01 -2.480390e-01 5 3.2145956903e-15 -2.21e+12 1.94e-04 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.499996e-01 6 1.2915932826e-24 -2.49e+14 1.94e-05 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2994882026e-24 6.08e+02 1.94e-04 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2994882026e-24 6.08e+02 1.94e-03 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2994882026e-24 6.08e+02 1.94e-02 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 7 1.2628874232e-24 -2.27e+03 1.94e-03 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 8 1.1896282775e-24 -6.16e+03 1.94e-04 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2110348070e-24 1.77e+03 1.94e-03 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2110348070e-24 1.77e+03 1.94e-02 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2110348070e-24 1.77e+03 1.94e-01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2219968181e-24 2.65e+03 1.94e+00 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2252721686e-24 2.91e+03 1.94e+01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 9 1.1888204839e-24 -6.79e+01 1.94e+00 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2998029581e-24 8.54e+03 1.94e+01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 10 1.1888204839e-24 0.00e+00 1.94e+00 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 iter chisq delta/lim lambda c1 c2 c3 c4 c5 c6 After 10 iterations the fit converged. final sum of squares of residuals : 1.18882e-24 rel. change during last iteration : 0 degrees of freedom (FIT_NDF) : 14 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 2.91403e-13 variance of residuals (reduced chisquare) = WSSR/ndf : 8.49157e-26 p-value of the Chisq distribution (FIT_P) : 1 Final set of parameters Standard Deviation ======================= ========================== c1 = 2.5 +/- 1.079 (43.17%) c2 = -1.8 +/- 0.9301 (51.67%) c3 = 70 +/- 0.2616 (0.3737%) c4 = -3.2 +/- 0.3759 (11.75%) c5 = 0.4 +/- 0.01302 (3.256%) c6 = -0.25 +/- 7.512 (3005%) correlation matrix of the fit parameters: c1 c2 c3 c4 c5 c6 c1 1.000 c2 0.124 1.000 c3 0.370 -0.129 1.000 c4 -0.440 -0.405 -0.182 1.000 c5 -0.447 0.062 -0.919 0.303 1.000 c6 0.945 0.197 0.366 -0.461 -0.428 1.000 --------------------------------------------------------- 6D fit: expected 2.5 c1 = 2.49999999999976 6D fit: expected -1.8 c2 = -1.80000000000026 6D fit: expected 70.0 c3 = 70.0 6D fit: expected -3.2 c4 = -3.19999999999975 6D fit: expected 0.4 c5 = 0.4 6D fit: expected -0.25 c6 = -0.250000000000491 FIT_NDF = 14 after range filters (expected 14) --------------------------------------------------------- Hit return to try fit with array variables"fitmulti.dem" line 183: warning: > Implied independent variable not found in fit function. > Assuming version 4 syntax with zerror in column 8 but no zerror keyword. iter chisq delta/lim lambda A[1] A[2] A[3] A[4] A[5] A[6] 0 3.4961271093e+05 0.00e+00 1.94e+01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1 7.6637354181e+04 -3.56e+05 1.94e+00 9.608331e-01 1.184762e+00 4.200546e+00 1.106839e+00 3.265869e+00 9.641673e-01 2 2.4111848689e+03 -3.08e+06 1.94e-01 1.293024e+00 2.647546e+00 5.815961e+01 -1.307965e+00 9.256806e-01 1.373154e-01 3 4.0425600039e-02 -5.96e+09 1.94e-02 2.493419e+00 -1.646308e+00 6.996638e+01 -3.211246e+00 4.014427e-01 -1.585899e-01 4 7.1103832630e-08 -5.69e+10 1.94e-03 2.500262e+00 -1.799903e+00 7.000002e+01 -3.200052e+00 3.999987e-01 -2.480390e-01 5 3.2145956903e-15 -2.21e+12 1.94e-04 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.499996e-01 6 1.2915932826e-24 -2.49e+14 1.94e-05 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2994882026e-24 6.08e+02 1.94e-04 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2994882026e-24 6.08e+02 1.94e-03 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2994882026e-24 6.08e+02 1.94e-02 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 7 1.2628874232e-24 -2.27e+03 1.94e-03 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 8 1.1896282775e-24 -6.16e+03 1.94e-04 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2110348070e-24 1.77e+03 1.94e-03 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2110348070e-24 1.77e+03 1.94e-02 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2110348070e-24 1.77e+03 1.94e-01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2219968181e-24 2.65e+03 1.94e+00 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2252721686e-24 2.91e+03 1.94e+01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 9 1.1888204839e-24 -6.79e+01 1.94e+00 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 * 1.2998029581e-24 8.54e+03 1.94e+01 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 10 1.1888204839e-24 0.00e+00 1.94e+00 2.500000e+00 -1.800000e+00 7.000000e+01 -3.200000e+00 4.000000e-01 -2.500000e-01 iter chisq delta/lim lambda A[1] A[2] A[3] A[4] A[5] A[6] After 10 iterations the fit converged. final sum of squares of residuals : 1.18882e-24 rel. change during last iteration : 0 degrees of freedom (FIT_NDF) : 14 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 2.91403e-13 variance of residuals (reduced chisquare) = WSSR/ndf : 8.49157e-26 p-value of the Chisq distribution (FIT_P) : 1 Final set of parameters Standard Deviation ======================= ========================== A[1] = 2.5 +/- 1.079 (43.17%) A[2] = -1.8 +/- 0.9301 (51.67%) A[3] = 70 +/- 0.2616 (0.3737%) A[4] = -3.2 +/- 0.3759 (11.75%) A[5] = 0.4 +/- 0.01302 (3.256%) A[6] = -0.25 +/- 7.512 (3005%) correlation matrix of the fit parameters: A[1] A[2] A[3] A[4] A[5] A[6] A[1] 1.000 A[2] 0.124 1.000 A[3] 0.370 -0.129 1.000 A[4] -0.440 -0.405 -0.182 1.000 A[5] -0.447 0.062 -0.919 0.303 1.000 A[6] 0.945 0.197 0.366 -0.461 -0.428 1.000 --------------------------------------------------------- Variables beginning with A_: A_1__err = 1.07920943632493 A_2__err = 0.930099132065864 A_3__err = 0.261566870454281 A_4__err = 0.375869300020404 A_5__err = 0.0130221281721936 A_6__err = 7.51152824151569 Array A after 6D fit: [2.49999999999976,-1.80000000000026,70.0,-3.19999999999975,0.4,-0.250000000000491] Hit return to end multidimension fit demo******************** file named_var.dem ******************** Hit return to continue******************** file param.dem ******************** Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue******************** file piecewise.dem ******************** Hit to continueHit to continueHit to continue******************** file polar.dem ******************** Hit return to continueHit return to continue"polar.dem" line 21: warning: Did you try to plot a complex-valued function? Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue******************** file poldat.dem ******************** Hit return to continueHit return to continueHit return to continueHit return to continue******************** file polargrid.dem ******************** polar mode is ON polar grid uses 36 theta wedges and 12 radial segments masked by theta range [-20:210] radial range [0:*] polar gridding scheme qnorm 1 set rrange [ * : * ] noreverse writeback noextend # (currently [0.00000:285.833] ) to continue to continue polar mode is ON polar grid uses 360 theta wedges and 50 radial segments masked by theta range [0:360] radial range [0:*] polar gridding scheme gauss kdensity scale 30 Theta increases clockwise with origin at top of plot to try logscale R to continue******************** file polar_quadrants.dem ******************** to continue******************** file sectors.dem ******************** to continue to continue to continue to continue to continue to continue to continue to continue to continue to continue to continue to continue to continue to continue******************** file orbits.dem ******************** ******************** file solar_path.dem **************** 21-12-2016 sunrise 8:04 sunset 16:37 sunlight 8 h 32 m 22-06-2017 sunrise 4:37 sunset 20:04 sunlight 15 h 27 m 31-12-2023 sunrise 8:03 sunset 16:38 sunlight 8 h 35 m to continue******************** file ttics.dem ******************** to continue to continue to continue******************** file boxplot.dem ******************** *** Boxplot demo *** Hit to continue: Compare sub-datasetsHit to continue: Assign selected colors to each factorHit to continue: Sort factors alphabeticallyHit to continue: The same, with iteration and manual filteringHit to continue: boxplot demo finished******************** file jitter.dem ******************** Hit to continueHit to continueHit to continueHit to continueHit to continueHit to continueHit to continue******************** file violinplot.dem ******************** Hit to continueHit to continueHit to continueHit to continueHit to continue******************** file spiderplot.dem ******************** to continue to continue to continue to continue to continue to continue to continue******************** file sampling.dem ******************** test 1: explicit trange distinct from xrange Hit to continuetest 2: range set by 'sample' keyword, linear x axis Hit to continuetest 3: range set by 'sample' keyword, logscale x axis Hit to continuetest 4: splot '++' with autoscaled y (linear xy) Hit to continuetest 5: splot '++' with autoscaled y (logscale xy) Hit to continuetest 6: plot '++' with image (linear xy) Hit to continuetest 7: plot '++' with image (logscale xy) Hit to continuetest 8: multiple sampling ranges in one 2D plot command Hit to continuetest 9: 3D sampling range distinct from plot x/y range Hit to continuetest 10: splot '++' with explicit sampling intervals Hit to continuetest 10: plot '++' with explicit sampling intervals Hit to continueHit return to continueHit return to continue******************** file multiplt.dem ******************** Hit return to continue to continue to continue to continue to continue to continue to continue******************** file surface1.dem ******************** Hit return to continueHit return to continue (1)Hit return to continue (2)Hit return to continue (3)Hit return to continue (4)Hit return to continue (5)Hit return to continue (6)Hit return to continue (7)Hit return to continue (8)Hit return to continue (9)Hit return to continue (10)Hit return to continue (11)Hit return to continue (12)Hit return to continue (13)Hit return to continue (14)Hit return to continue (15)Hit return to continue (16)Hit return to continue (17)Hit return to continue (18)Hit return to continue (19)Hit return to continue (20)Hit return to continue (21)Hit return to continue (22)Hit return to continue (23)Hit return to continue (24)Hit return to continue (25)******************** file surface_explicit.dem ******************** to continue to continue******************** file discrete.dem ******************** Hit return to continueHit return to continueHit return to continue******************** file hidden.dem ******************** Hit return to continue (1)Hit return to continue (2)Hit return to continue (3)Hit return to continue (4)Hit return to continue (5)Hit return to continue (6)Hit return to continue (7)******************** file hidden_compare.dem ******************** to continue******************** dgrid3d ******************** Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continuePress Return to continue - the plot may take some time to appearPress Return to continue - the plot may take some time to appearHit return to continue******************** file world.dem ******************** Hit return to continueHit return to continueHit return to continueHit return to continueSame plot with hidden line removalHit return to continue******************** file prob.dem ******************** Statistical Library Demo, version 2.3 Copyright (c) 1991, 1992, Jos van de Woude, jvdwoude@hut.nl Press Ctrl-C to exit right now Press Return to start demo ...Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue******************** file prob2.dem ******************** Hit return for inverse error function.Hit return for inverse normal distribution function.Press return to continue Statistical Approximations, version 1.1 Copyright (c) 1991, 1992, Jos van de Woude, jvdwoude@hut.nl Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue******************** file random.dem ******************** Hit return to continue3D plot ahead, one moment please ... Hit return to continue Multivariate normal distribution The surface plot shows a two variable multivariate probability density function. On the x-y plane are some samples of the random vector and a contour plot illustrating the correlation, which in this case is zero, i.e. a circle. (Easier to view in map mode.) "random.dem" line 73: warning: Cannot contour non grid data. Please use "set dgrid3d". Hit return to continue Simple Monte Carlo simulation The first curve is a histogram where the binned frequency of occurrence of a pseudo random variable distributed according to the normal (Gaussian) law is scaled such that the histogram converges to the normal probability density function with increasing number of samples used in the Monte Carlo simulation. The second curve is the normal probability density function with unit variance and zero mean. Hit return to continue Another Monte Carlo simulation This is similar to the previous simulation but uses multivariate zero mean, unit variance normal data by computing the distance each point is from the origin. That distribution is known to fit the Maxwell probability law, as shown. Hit return to continue******************** file rugplot.dem ******************** Hit to continue******************** file smooth.dem ******************** Hit enter to continueHit enter to continueHit enter to continueHit enter to continue******************** file spline.dem ******************** Press return to continuePress return to continuePress return to continuePress return to continuePress return to continuePress return to continue******************** file sharpen.dem ******************** to continue******************** file binary.dem ******************** Hit return to continue (1)Hit return to continue (2)Hit return to continue (3)******************** file steps.dem ******************** Hit return for demonstration of automatic histogram creationHit return to see the same plot with fillstepsPress return to continue******************** file scatter.dem ******************** Hit return to continue (1)Hit return to continue (2)Hit return to continue (3)Hit return to continue (4)Hit return to continue (5)"scatter.dem" line 43: warning: Cannot contour non grid data. Please use "set dgrid3d". Hit return to continue (6)Hit return to continue (7)Hit return to continue (8)******************** file singulr.dem ******************** Hit return to continue (1)Hit return to continue (2)Hit return to continue (3)Hit return to continue (4)Hit return to continue (5)Hit return to continue (6)Hit return to continue (7)Hit return to continue (8)Hit return to continue (9)Hit return to continue (10)Hit return to continue (11)Hit return to continue (12)Hit return to continue (13)Hit return to continue (14)Hit return to continue (15)Hit return to continue (16)Hit return to continue (17)Hit return to continue (18)Hit return to continue (19)Hit return to continue (20)******************** file airfoil.dem ******************** NACA four series airfoils by bezier splines Will add pressure distribution later with Overplotting Press ReturnPress ReturnJoukowski Airfoil using Complex Variables Press ReturnPress Return******************** file surface2.dem ******************** Hit return to continue (1)Hit return to continue (2)Hit return to continue (3)Hit return to continue (4)Hit return to continue (5)Hit return to continue (6)Hit return to continue (7)Hit return to continue (8)Hit return to continue (9)******************** file azimuth.dem ******************** Hit return to continue******************** file projection.dem ****************** Hit return to continue******************** contours ******************** Hit return to continue (1)Hit return to continue (2)Hit return to continue (3)Hit return to continue (4)Hit return to continue (5)Hit return to continue (6)Hit return to continue (7)Hit return to continue (8)Hit return to continue (9)Hit return to continue (10)Hit return to continue (11)Hit return to continue (12)Hit return to continue (13)Hit return to continue (14)Hit return to continue (15)Hit return to continue (16)Hit return to continue (17)Hit return to continue (18)Hit return to continue (19)Hit Return to Continue (20)Hit Return to Continue (21)Hit Return to Continue (22)Hit Return to Continue (23) to continue******************** file contourfill.dem ******************** to continue to set contourfill ztics for 2D projection to continue******************** file pixmap.dem ******************** Hit to continue******************** file bivariat.dem ******************** Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue******************** Time/Date data ******************** Hit return to continueHit return to continueHit return to continue Relative time output (strftime(), axis labels) t = -3672.5 print strftime("%.2tM == %.2tS", t) -61.21 min == -3672.50 sec print strftime("%tM:%.2tS", t) -61:12.50 print strftime("%tH:%tM:%.2tS", t) -1:01:12.50 t = 3672.5 print strftime("%.2tM == %.2tS", t) 61.21 min == 3672.50 sec print strftime("%tM:%.2tS", t) 61:12.50 print strftime("%tH:%tM:%.2tS", t) 1:01:12.50 Relative time input (strptime(), data files) print strptime("%tH:%tM:%tS", "-1:01:12.50") -3672.5 print strptime(" %tM:%tS", "-61:12.50") -3672.5 print strptime(" %tS", "-3672.50") -3672.5 Timezones time output (strftime(), axis labels) t = 1550496278 print strftime("%d/%m/%y\t%H:%M", t) 18/02/19 13:24 print strftime("%d/%m/%y\t%H:%M%z", t) 18/02/19 13:24 print strftime("%d/%m/%y\t%H:%M%Z", t) 18/02/19 13:24 Timezones time input (strptime(), data files) print strptime("%d/%m/%y\t%H:%M", "18/02/19\t13:24") 1550496240.0 print strptime("%d/%m/%y\t%H:%M%z", "18/02/19\t12:24+00:00") "timedat.dem" line 89: warning: Bad time format %z 1550492640.0 print strptime("%d/%m/%y\t%H:%M%z", "18/02/19\t13:24+01:00") "timedat.dem" line 90: warning: Bad time format %z 1550496240.0 print strptime("%d/%m/%y\t%H:%M %Z", "18/02/19\t13:24 CET") "timedat.dem" line 91: warning: Bad time format %Z 1550496240.0 print strptime("%d/%m/%y\t%H:%M %Z", "18/02/19\t14:24 CEST") "timedat.dem" line 92: warning: Bad time format %Z 1550499840.0 Hit return to check backwards compatibility with v4 syntaxHit return to continue********************** file rainbow.dem ********************* # These are the input commands set style line 1 lt rgb "red" lw 3 set style line 2 lt rgb "orange" lw 2 set style line 3 lt rgb "yellow" lw 3 set style line 4 lt rgb "green" lw 2 set style line 5 lt rgb "cyan" lw 3 set style line 6 lt rgb "blue" lw 2 set style line 7 lt rgb "violet" lw 3 # And this is the result linestyle 1, linecolor rgb "red" linewidth 3.000 dashtype solid pointtype 1 pointsize default linestyle 2, linecolor rgb "orange" linewidth 2.000 dashtype solid pointtype 2 pointsize default linestyle 3, linecolor rgb "yellow" linewidth 3.000 dashtype solid pointtype 3 pointsize default linestyle 4, linecolor rgb "green" linewidth 2.000 dashtype solid pointtype 4 pointsize default linestyle 5, linecolor rgb "cyan" linewidth 3.000 dashtype solid pointtype 5 pointsize default linestyle 6, linecolor rgb "blue" linewidth 2.000 dashtype solid pointtype 6 pointsize default linestyle 7, linecolor rgb "violet" linewidth 3.000 dashtype solid pointtype 7 pointsize default Hit return to continueHit return to continue********************** file rgb_variable.dem ********************* Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue********************** file rgba_lines.dem ********************* Hit return to continue********************** file varcolor.dem ********************* Hit to continueHit to continueHit to continueHit to continueHit to continueHit to continueHit to continueHit to continue********************** file pt_variable.dem ********************* to continue to continue to continue********************** file pm3d.dem ********************* Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continuePress Enter; I will continue by 'set autoscale cb' and much more...Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continuePlot by pm3d algorithm draws quadrangles filled with color calculated from the z- or color-value of the surrounding 4 corners. The following demo shows different color spots for a plot with very small number of quadrangles (here rectangular pixels). Note that the default option is 'mean'. Hit return to continueEnd of pm3d demo. ********************** file pm3d_clip.dem ********************* to continue to continue********************** file complex_trig.dem ********************* ********************** libcerf routines ************************* This copy of gnuplot was not linked against libcerf This copy of gnuplot was not linked against libcerf This copy of gnuplot was not linked against libcerf ********************** libamos routines ************************* This copy of gnuplot does not support Ai, Bi This copy of gnuplot does not support BesselK to continueThis copy of gnuplot does not support complex expint ********************** special functions ************************* to continue to continue to continue to continue to continue to continue to continueHit return to continueHit return to continueHit return to continueHit return to continue to continue********************** file heatmaps ********************* Hit return to continueHit return to continueHit return to continueLoaded 14 points into 5 x 5 sparse matrix Hit return to continueLoaded 14 points into 5 x 5 sparse matrix Hit return to continueHit return to continueHit return to continue to use a finer grid to continue to continue********************** file matrix_index.dem ********************* Hit return to continue********************** file matrix_every.dem ********************* to continue********************** file pm3dgamma.dem ********************* Hit return to continue********************** file hidden2.dem *********************** Hit return to continueHit return to continueHit return to continue********************** file textcolor.dem ********************* Hit return to continueHit return to continue********************** file textrotate.dem ********************* Hit return to continue********************** enhanced text ********************* Hit return to continueHit return to continue********************** unicode text ********************* to continue********************** file dashtypes.dem ********************* Hit return to continueHit return to continue********************** file arrowstyle.dem ********************* We have defined the following arrowstyles: arrowstyle 1, head back linecolor rgb "dark-violet" linewidth 2.000 dashtype solid arrow heads: filled, length (screen units) 0.025, angle 30 deg, backangle 45 deg arrowstyle 2, head back linecolor rgb "#56b4e9" linewidth 2.000 dashtype solid arrow heads: nofilled, length (screen units) 0.03, angle 15 deg arrowstyle 3, head back linecolor rgb "dark-violet" linewidth 2.000 dashtype solid arrow heads: filled, length (screen units) 0.03, angle 15 deg, backangle 45 deg arrowstyle 4, head back linecolor rgb "#56b4e9" linewidth 2.000 dashtype solid arrow heads: filled, length (screen units) 0.03, angle 15 deg, backangle 90 deg arrowstyle 5, heads back linecolor rgb "dark-violet" linewidth 2.000 dashtype solid arrow heads: noborder, length (screen units) 0.03, angle 15 deg, backangle 135 deg arrowstyle 6, head back linecolor rgb "#56b4e9" linewidth 2.000 dashtype solid arrow heads: empty, length (screen units) 0.03, angle 15 deg, backangle 135 deg arrowstyle 7, nohead back linecolor rgb "dark-violet" linewidth 2.000 dashtype solid arrowstyle 8, heads back linecolor rgb "#56b4e9" linewidth 2.000 dashtype solid arrow heads: nofilled, length (screen units) 0.008, angle 90 deg Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHi return to continue********************** file vector.dem ********************* This file demonstrates -1- saving contour lines as a gnuplottable datablock -2- plotting a vector field on the same graph -3- manipulating columns using the '$1,$2' syntax. the example is taken here from Physics is the display of equipotential lines and electrostatic field for a dipole (+q,-q) Now create a in-memory datablock with equipotential lines Hit return to continueNow create a x/y datablock for plotting with vectors and display vectors parallel to the electrostatic field Hit return to continue"vector.dem" line 85: warning: Warning - difficulty fitting plot titles into key Hit return to continue********************** file arrows.dem ********************* Hit to continueHit to continue********************** file short_vector.dem ********************* to continue********************** file tics.dem ********************* Hit return to continueHit return to continueHit return to continueHit return to continueEnd of tics demo. ********************** file break_continue.dem ********************* start 1 continue if i == 3, break if i > 4 still in load file end 1 start 2 continue if i == 3, break if i > 4 still in load file end 2 start 3 continue if i == 3, break if i > 4 start 4 continue if i == 3, break if i > 4 still in load file end 4 start 5 continue if i == 3, break if i > 4 done with for loop ********************** file callargs.dem ********************* Entering callargs.dem with 0 parameters Now exercise the call mechanism at line 47 Entering callargs.dem with 8 parameters Test whether this copy of gnuplot also supports deprecated call parameter syntax $0 $1 $2 etc: no Variables beginning with ARG: ARGC = 8 ARGV = <8 element array> ARG0 = "callargs.dem" ARG1 = "1.23e4" ARG2 = "string constant" ARG3 = "FOO" ARG4 = "5.67" ARG5 = "3 + log(BAZ)" ARG6 = "a string" ARG7 = "8" ARG8 = "3.14159" ARG9 = "" ARG1 (numerical constant) came through as 1.23e4 @ARG1 = 12300.0 (ARG1 == @ARG1) is TRUE ARG2 (string constant) came through as string constant words(ARG2) = 2 ARG3 (undefined variable FOO) came through as FOO ARG4 (numerical variable BAZ=5.67) came through as 5.67 @ARG4 = 5.67 ARG5 (quoted expression) came through as 3 + log(BAZ) @ARG5 = 4.73518911773966 ARG6 (string variable) came through as a string words(ARG6) = 2 ARG7 (expression) came through as 8 ARG8 (pi) came through as 3.14159 ARGV = [12300.0,"string constant","FOO",5.67,"3 + log(BAZ)","a string",8,3.14159265358979] ********************** file volatile.dem ********************* ********************** file datastrings.dem ********************* to plot again using x2ticlabels to plot again using x2ticlabels to plot same data from table format to show double use of y values to show use of boxed labels to end demo********************** file textbox.dem ********************* to continue********************** file hypertext.dem ********************* hit return to continue********************** file rotate_labels.dem ***************** to continue********************** file stats.dem ********************* * FILE: Records: 20 Out of range: 0 Invalid: 0 Header records: 0 Blank: 1 Data Blocks: 1 * COLUMN: Mean: 2.5408 Std Dev: 0.2227 Sample StdDev: 0.2285 Skewness: 0.9783 Kurtosis: 3.6056 Avg Dev: 0.1785 Sum: 50.8156 Sum Sq.: 130.1035 Mean Err.: 0.0498 Std Dev Err.: 0.0352 Skewness Err.: 0.5477 Kurtosis Err.: 1.0954 Minimum: 2.2009 [ 7] Maximum: 3.1397 [ 9] Quartile: 2.3866 Median: 2.4610 Quartile: 2.6562 Hit return to continue * FILE: Records: 20 Out of range: 0 Invalid: 0 Header records: 0 Blank: 1 Data Blocks: 1 * COLUMNS: Mean: 2.1168 2.5408 Std Dev: 0.7921 0.2227 Sample StdDev: 0.8127 0.2285 Skewness: -1.1174 0.9783 Kurtosis: 3.5250 3.6056 Avg Dev: 0.6330 0.1785 Sum: 42.3356 50.8156 Sum Sq.: 102.1648 130.1035 Mean Err.: 0.1771 0.0498 Std Dev Err.: 0.1252 0.0352 Skewness Err.: 0.5477 0.5477 Kurtosis Err.: 1.0954 1.0954 Minimum: 0.0000 [ 0] 2.2009 [ 7] Maximum: 2.9957 [19] 3.1397 [ 9] Quartile: 1.7006 2.3866 Median: 2.3502 2.4610 Quartile: 2.7403 2.6562 Linear Model: y = -0.06694 x + 2.682 Slope: -0.06694 +- 0.06437 Intercept: 2.682 +- 0.1455 Correlation: r = -0.2381 Sum xy: 106.7 Hit return to continue********************** file iterate.dem ********************* Hit return to continueHit return to continueHit return to continue dynamic reevaluation of numeric iteration limits J = [1,4,3] do for [i=1:3] for [j=J[i]:3] { save(i,j) } 1-1 1-2 1-3 3-3 J = [4,1,3] do for [i=1:3] for [j=J[i]:3] { save(i,j) } 2-1 2-2 2-3 3-3 do for [i=1:4] for [k=i:i] for [j=1:k] { save(i,j) } 1-1 2-1 2-2 3-1 3-2 3-3 4-1 4-2 4-3 4-4 dynamic reevaluation of iteration string A = ["a b","","d"] do for [ i = 1:|A| ] { do for [ j in A[i]] { print "".i.": ".j }} 1: a 1: b 3: d do for [ i = 1:|A| ] for [ j in A[i]] { print "".i.": ".j } 1: a 1: b 3: d ********************** histograms ********************* to plot the same data as a histogram to change the gap between clusters to plot the same dataset as stacked histogram to rescale each stack to % of totalNow try histograms stacked by columnsNext we do several sets of parallel histogramsSame plot using rowstacked histogram to finish histogram demoSame plot using explicit histogram start colorsSame plot using explicit histogram start patternSame plot with both explicit color and patternHit return to continue to continue to continue to continue********************** file boxclusters.dem ********************* to continue********************** Array functions ********************* Sum[ 5] = 132551 Sum[ 6] = 218066 Sum[ 7] = 363446 Sum[ 8] = 2441994 Sum[ 9] = 570452 Sum[10] = 1191986 Sum[11] = 4787900 Sum[12] = 250579 Sum[13] = 473705 to continue to fit function to array valuesiter chisq delta/lim lambda a b c 0 1.0028328575e+02 0.00e+00 2.34e-01 1.000000e-02 1.000000e-02 1.000000e-02 1 1.5801742112e+01 -5.35e+05 2.34e-02 1.606160e-02 3.087431e-02 -1.313622e-02 2 1.4408428171e+01 -9.67e+03 2.34e-03 -9.632124e-02 3.471172e-02 9.547984e-02 3 8.1377220374e+00 -7.71e+04 2.34e-04 -6.860755e-01 4.691257e-02 8.742445e-02 4 1.2433205970e+00 -5.55e+05 2.34e-05 -1.820671e+00 6.718158e-02 -2.845982e-03 5 6.1954576313e-02 -1.91e+06 2.34e-06 -1.591434e+00 6.308333e-02 4.504819e-02 6 6.0984370182e-02 -1.59e+03 2.34e-07 -1.584251e+00 6.295384e-02 4.625961e-02 7 6.0984340867e-02 -4.81e-02 2.34e-08 -1.584224e+00 6.295329e-02 4.625481e-02 iter chisq delta/lim lambda a b c After 7 iterations the fit converged. final sum of squares of residuals : 0.0609843 rel. change during last iteration : -4.80707e-07 degrees of freedom (FIT_NDF) : 97 rms of residuals (FIT_STDFIT) = sqrt(WSSR/ndf) : 0.025074 variance of residuals (reduced chisquare) = WSSR/ndf : 0.000628705 Final set of parameters Asymptotic Standard Error ======================= ========================== a = -1.58422 +/- 0.006802 (0.4294%) b = 0.0629533 +/- 0.0001141 (0.1812%) c = 0.0462548 +/- 0.002508 (5.422%) correlation matrix of the fit parameters: a b c a 1.000 b -0.854 1.000 c 0.015 -0.020 1.000 to continue * FILE: Records: 20 Out of range: 0 Invalid: 0 Header records: 0 Blank: 1 Data Blocks: 1 * COLUMN: Mean: 5.9866 Std Dev: 5.5330 Sample StdDev: 5.6768 Skewness: 1.4119 Kurtosis: 3.1850 Avg Dev: 4.3558 Sum: 119.7324 Sum Sq.: 1329.0836 Mean Err.: 1.2372 Std Dev Err.: 0.8749 Skewness Err.: 0.5477 Kurtosis Err.: 1.0954 Minimum: 1.0000 [ 8] Maximum: 17.8341 [ 0] Quartile: 3.2693 Median: 3.5672 Quartile: 4.8706 * FILE: Records: 20 Out of range: 0 Invalid: 0 Header records: 0 Blank: 1 Data Blocks: 1 * COLUMN: Mean: 5.9866 Std Dev: 5.5330 Sample StdDev: 5.6768 Skewness: 1.4119 Kurtosis: 3.1850 Avg Dev: 4.3558 Sum: 119.7324 Sum Sq.: 1329.0836 Mean Err.: 1.2372 Std Dev Err.: 0.8749 Skewness Err.: 0.5477 Kurtosis Err.: 1.0954 Minimum: 1.0000 [ 8] Maximum: 17.8341 [ 0] Quartile: 3.2693 Median: 3.5672 Quartile: 4.8706 to continueA[ 1 ] = 1 A[ 1 :6] = [1,2,3.0,4.0,"five","six"] A[ 2 ] = 2 A[ 2 :6] = [2,3.0,4.0,"five","six"] A[ 3 ] = 3.0 A[ 3 :6] = [3.0,4.0,"five","six"] A[ 4 ] = 4.0 A[ 4 :6] = [4.0,"five","six"] A[ 5 ] = five A[ 5 :6] = ["five","six"] A[ 6 ] = six A[ 6 :6] = ["six"] A[ 7 ] = {0.0, 7.0} A[ 7 :6] = [] A[ 8 ] = {8.0, 8.0} A[ 8 :6] = [] no member of A matches NaN array B = [2,3.0,4.0] Key/value pairs water is blue dirt is brown sky is blue split: OK join: OK ********************** Image formats ********************* The plotting styles `image` and `rgbimage` are intended for plotting images described in a data file either in the conventional ASCII format or in a binary format described by the qualifiers `binary` and `using`. All pixels have an (x,y) or (x,y,z) coordinate. These values can be included in the data file or implicitly determined with the sampling 'array' key word and sampling periods 'dx' and 'dy'. The key words 'rotate' and, for 3d plots, 'perpendicular' control orientation. The data for this image was stored as RGB triples, one byte per channel, without (x,y) coordinate information. This yields a most compact file. The plotting command is displayed on the graph. Hit return to continue Images are typically stored in a file with the first datum being the top, left pixel. Without the ability to translate coordinates, the the result would be an upside down image. The key word 'array' means an implied sample array is applied to generate the locations of file data using the sampling periods dx, dy and dz. The x-dimension is always the contiguous points in a binary file. The y-dimension is the line number which is incremented upon the x-dimension reaching the line length. The z-dimension is the plane number which is incremented upon the y-dimension reaching the number of lines per plane. To alter the location of the binary data when displayed via the 'plot' command, use the key word 'rotate' along with changing the sign of dx, dy and dz. Hit return to continue There is the ability to plot both color images and palette based images. This is controlled by the styles `image`, which derives color information from the current palette, and `rgbimage`, which requires three components representing the red, blue and green primary color scheme. By the way, if you have a mouse active, click the right button inside the image to isolate a portion of the image and see what happens. Hit return to continue Naturally, as with 3d color surfaces, the palette may be changed. This is an example of gray scale. Also, notice in the plot command the key word 'flipy'. This means to change the direction of the sample along the y dimension and is useful for the situation where images or other data are stored in some direction other than that of the Cartesion system. Alone, 'flipD' means do flipping in the D (x y or z) direction for all records. Individual records can be controlled using the syntax 'flipD=#,...,#', where # is '0' or '1'. Hit return to continue Also, similar to 3d color surface plots, a color box showing the palette mapping scheme can be added to the plot. The default location is the right edge of the plot. The location can be set manually using `set colorbox` and `set margin`. As a prelude to the next graph, resize the plot window to judge the refresh speed of the image drawing routine. Notice that when the window is smaller, the image refresh is faster. There is more decimation in the data of the original image and less data to plot. Furthermore, the window continues to refresh at a reasonable rate even when the input image size becomes large (e.g., 1024 x 1024) because the number of pixels on the screen remains about the same while much of the hi resolution data is decimated. Hit return to continue The 'rotation' key word works not only with angles of integer multiples of 90 degrees but also arbitrary rotations. When constructing an image, Gnuplot verifies that pixel locations form a valid grid. Pixel widths are based upon the grid spacing. If the image orientation is aligned with the view axes, Gnuplot uses an efficient image driver routine. Otherwise, individual pixels are drawn using polygon shapes. Resize this window and compare the plot's refresh rate to that of the previous and next plot. Notice how in this example if the window is small the image refresh does not speed up. Unlike the image routine where image data is decimated, all color rectangles must be redrawn no matter the size of the output image. Also notice how the center of the image matches the tuple specified with the key word 'center' in the plot command. Before doing the rotation, the image's natural center is subtracted, and after doing the rotation, the specified center is added. Hit return to continue The image of this plot is rotated 90 degrees and can utilize the efficient image functions of terminal drivers. The plot refresh is faster than the previous plot. Furthermore, the image location in this case is specified via the 'origin' key word. As such, the rotation is done about the origin as opposed to the center of the image. Notice the difference in axis ranges. Hit return to continue Algebraic manipulation of the input variables can select various components of the image. Here are three examples where two channels--or analogous to the ASCII file, data "columns"--are ignored This is done by using `*` in the format to indicate that a variable of a certain size should be discarded. For example, to select the green channel, `%*uchar%uchar%*uchar` is one alternative. Hit return to continue The range of valid RGB color component values is [0:255] This is a CHANGE in gnuplot version 5.2. To adjust the color balance you can filter the individual values through a scaling function. Here we multiply by a constant c, c > 1 to brighten, c < 1 to dim. Hit return to continue Not only can the 2d binary data mode be used for image data. Here is an example that repeats the `using.dem` demo with the same data, but stored in binary format of differing sizes. It uses different format specifiers within the 'format' string. There are machine dependent and machine independent specifiers, display by the command 'show datafile binary datasizes': The following binary data sizes are machine dependent: name (size in bytes) "char" "schar" "c" (1) "uchar" (1) "short" (2) "ushort" (2) "int" "sint" "i" "d" (4) "uint" "u" (4) "long" "ld" (8) "ulong" "lu" (8) "float" "f" (4) "double" "lf" (8) (8) (8) The following binary data sizes attempt to be machine independent: name (size in bytes) "int8" "byte" (1) "uint8" "ubyte" (1) "int16" "word" (2) "uint16" "uword" (2) "int32" (4) "uint32" (4) "int64" (8) "uint64" (8) "float32" (4) "float64" (8) Hit return to continue Again, a different format specification for `using` can be used to select different "columns" within the file. Hit return to continue Here is another example, one repeating the `scatter.dem` demo. With binary data we cannot have blank lines to indicate a break in data, as is done with ASCII files. Instead, we can specify the record lengths in the command. In this case, the data file contains the (x,y,z) coordinate information, hence implicit derivation of that information is not desired. Instead, the record lengths can be specified using the keyword 'record', which behaves the same as 'array' but does not generate coordinates. The command is displayed on the graph. Hit return to continue For binary data, the byte endian format of the file and of the compiler often require attention. Therefore, the key word 'endian' is provided for setting or interchanging byte order. The allowable types are 'little', 'big', and depending upon how your version of Gnuplot was compiled, 'middle' (or 'pdp') for those still living in the medieval age of computers. These types refer to the file's endian. Gnuplot arranges bytes according to this endian and what it determines to be the compiler's endian. There are also the special types 'default' and 'swap' (or 'swab') for those who don't know the file type but realize their data looks incorrect and want to change the byte read order. Here is an example showing the `scatter.dem` data plotted with correct and incorrect byte order. The file is known to be little endian, so the upper left plot is correct appearance and the upper right plot is incorrect appearance. The lower two plots are default and swapped bytes. If the plots within the columns match, your compiler uses little endian. If diagonal plots match then your compiler uses big endian. If neither of the bottom plots matches the upper plots, Tux says you're living in the past. Hit return to continue This close up of a 2x2 image illustrates how pixels surround the sampling grid points. This behavior is slightly different than that for pm3d where the four grid points would be used to create a single polygon element using an average, or similar mathematical combination, of the four values at those points. Hit return to continue Lower dimensional data may be extended to higher dimensional plots via a number of simple, logical rules. The first step Gnuplot does is sets the components for higher than the natural dimension of the input data to zero. For example, a two dimensional image can be displayed in the three dimensional plot as shown. Without translation, the image lies in the x/y-plane. Warning: empty z range [0:0], adjusting to [-1:1] Hit return to continue The key words 'rotate' and 'center' still apply in 'splot' with rules similar to their use in 'plot'. However, the center must be specified as a three element tuple. Warning: empty z range [50:50], adjusting to [49.5:50.5] Hit return to continue To have full degrees of freedom in orienting the image, an additional key word, 'perpendicular', can translate the x/y-plane of the 2d data so that it lies orthogonal to a vector given as a three element tuple. The default vector is, of course, (0,0,1). The vector need not be of unit length, as this example illustrates. Viewing this plot with the mouse active can help visualize the image's orientation by panning the axes. Hit return to continue These concepts of extending lower dimensional data also apply to temporal-like signals. For example, a uniformly sampled sinusoid, sin(1.75*pi*x), in a binary file having no data for the independent variable can be displayed along any direction for both 'plot'... Hit return to continue ...and 'splot'. Here is the 'scatter.dem' example again, but this simulates the case of the redundant x coordinates not being in the binary file. The first "column" of the binary file is ignored and reconstructed by orienting the various data records. Hit return to continue Some binary data files have headers, which may be skipped via the 'skip' key word. Here is the 'scatter.dem' example again, but this time the first and third traces are skipped. The first trace is 30 samples of three floats so takes up 360 bytes of space. Similarly, the third trace takes up 348 bytes. Hit return to continue Generating uniformly spaced coordinates is valid for polar plots as well. This is useful for data acquired by machines sampling in a circular fashion. Here the sinusoidal data of the previous 2D plot put on a polar plot. Note the pseudonyms 'dt' meaning sample period along the angular, or theta, direction. In Gnuplot, cylindrical coordinate notation is (t,r,z). [Different from common math convention (r,t,z).] Hit return to continue Binary data stored in matrix format (i.e., gnuplot binary) may also be translated with similar syntax. However, the binary keywords `format`, `array` and `record` do not apply because gnuplot binary has the requirements of float data and grid information as part of the file. Here is an example of a single matrix binary file input four times, each translated to a different location with different orientation. Hit return to continue As with ASCII data, decimation in various directions can be achieved via the `every` keyword. (Note that no down- sampling filter is applied such that you risk aliasing data with the `every` keyword. Here is a series of plots with increasing decimation. Hit return to continueHit return to continueHit return to continueHit return to continue Decimation works on general binary data files as well. Here is the image file with increasing decimation. Hit return to continueHit return to continue Gnuplot understands a few common binary formats. Internally a function is linked with various extensions. When the extension is specified at the command line or recognized via a special file type called 'auto', Gnuplot will call the function that sets up the necessary binary information. The known extensions are displayed using the 'show filetype' command. E.g., This version of gnuplot understands the following binary file types: avs bin edf ehf gif gpbin jpeg jpg png raw rgb auto Here's an example where an EDF file is recognized when Gnuplot is in 'auto' mode. Details are pulled from the header of file itself and not specified at the command line. The command line can still be used to over-ride in-file attributes. Hit return to continue The 'flip', 'rotate' and 'perpendicular' qualifiers should provide adequate freedom to orient data as desired. However, there is an additional key words 'scan' which may offer a more direct and intuitive manner of orienting data depending upon the user's application and perspective. 'scan' is a 2 or 3 letter string representing how Gnuplot should derive (x,y), (x,y,z), (t,r) or (t,r,z) from the the datafile's scan order. The first letter pertains to the fastest rate or point-by-point increment. The second letter pertains to the medium rate or line-by-line increment. If there is a third letter, it pertains to the slowest rate or plane-by-plane increment. The default or inherent scan order is 'scan=xyz'. The pseudonym 'transpose' is equivalent to 'scan=yx' when generating 2D coordinates and 'scan=yxz' when generating 3D coordinates. There is a subtle difference between the behavior of 'scan' when dimension info is taken from the file itself as opposed to entered at the command line. When information is gathered from the file, internal scanning is unaltered so that issuing the 'scan' command may cause the number of samples along the various dimensions to change. However, when the qualifier 'array' is entered at the command line, the array dimensions adjust so that 'array=XxYxZ' is always the number of samples along the Cartesian x, y and z directions, respectively. Hit return to continue It is possible to enter binary data at the command line. Of course, the limitation to this approach is that keyboards will allow entering only a limited subset of the possible character values necessary to represent general binary data. For this reason, the primary application for binary data at the command line is using Gnuplot through a pipe. For example, if a pipe is established with a C program, the function 'fputs()' can send ASCII strings containing the Gnuplot commands while the function 'fwrite()' can send binary data. Furthermore, there can be no special ending character such as in the case of ASCII data entry where 'e' represents the end of data for the special file '-'. It is important to note that when 'binary' is specified, Gnuplot will continue reading until reaching the number of elements specified via the 'array' or 'record' command. Here is an example of binary data in the range [0:1] inserted into the command stream by copying 48 bytes from a pre-existing binary file into this demo file. Hit return to continue ASCII data files have a matrix variant. Unlike matrix binary, ASCII binary may have multiple matrices per file, separated by a blank line. The keyword `index` can select the desired matrix to plot. Hit return to continue Images maintain orientation with respect to axis direction. All plots show the same exact plot, but with various states of reversed axes. The upper left plot has reversed x axis, the upper right plot has conventional axes, the lower left plot has both reversed x and y axes, and the lower right plot has reversed y axis. Hit return to continue Tux says "bye-bye". Hit return to continue End of image demo... "imageNaN.dem" line 38: warning: matrix contains missing or undefined values Hit return to continue"imageNaN.dem" line 45: warning: matrix contains missing or undefined values Hit return to continue"imageNaN.dem" line 51: warning: matrix contains missing or undefined values Hit return to continue"imageNaN.dem" line 57: warning: matrix contains missing or undefined values Hit return to continue"imageNaN.dem" line 63: warning: matrix contains missing or undefined values Hit return to continue"imageNaN.dem" line 69: warning: matrix contains missing or undefined values Hit return to continueHit return to continueHit return to continueHit return to continue********************** file stringvar.dem ********************* Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue time_str = " 2005-05-09 19:44:12 " -> seconds = 1115667852.0 seconds + 10. = 1115667862.0 -> time_str2 = " 2005-05-09 19:44:22 " read_time(fmt, c) = strptime(fmt, stringcolumn(c).' '.stringcolumn(c+1)) Hit return to continue********************** file running_avg.dem ********************* Hit return to continue********************** file pointsize.dem ********************* Hit return to continueHit return to continueHit return to continueHit return to continue********************** file circles.dem ********************* Hit return to continueHit return to continue********************** file armillary.dem ********************* Hit to continue********************** file ellipses_style.dem ********************* Hit to continueHit to continueHit to continueHit to continueHit to continueHit to continueHit to continueHit to continueHit to continueHit to continueHit to continue********************** file key.dem ********************* Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue********************** custom key layout ********************* Hit return to continueHit return to continueHit return to continue to continue to continue********************** file walls.dem ********************* to continue********************** file boxes3d.dem ********************* hit return to continuehit return to continuehit return to continuehit return to continuehit return to continuehit return to continue********************** file borders.dem ********************* Hit return to continue********************** file columnhead.dem ********************* ********************** file margins.dem ********************* Hit return to continue********************** file rectangle.dem ********************* Hit return to continueHit return to continue********************** clipping ********************* Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Hit return to continueHit return to continueHit return to continue to continue********************** file approximate.dem ********************* Hit return to continue********************** file parallel.dem ********************* Hit return to continueHit return to continueHit return to continue********************** nonlinear axis demos ********************* to continue to continue to continue to continue to continue to continue to continue to continue********************** linked axes ************************** Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continue********************** file map_projection.dem ********************* to continue to continue to continue********************** file transparent.dem ********************* Hit return to continueHit return to continueHit return to continueHit return to continue********************** file transparent_solids.dem ********************* Hit return to continueHit return to continue********************** file pm3d_lighting.dem ********************* Hit to continueHit return to continue********************** file polygons.dem ********************* Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Polygon is not closed - adding extra vertex Hit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continueHit return to continue to continue********************** file named_palettes.dem ********************* to continue to continue to continue to continue********************** file palette+alpha.dem ********************* Hit return to continue********************** file argb_hexdata.dem ********************* to continueWarning: empty z range [1:1], adjusting to [0.99:1.01] to continue********************** file vplot.dem ********************* vfill from - : radius 1 gives a brick of 25 voxels on x, 25 voxels on y, 25 voxels on z number of points input: 1 number of voxels modified: 7901 vfill from - : radius 1 gives a brick of 25 voxels on x, 25 voxels on y, 25 voxels on z number of points input: 1 number of voxels modified: 7930 vfill from - : radius 2 gives a brick of 19 voxels on x, 19 voxels on y, 19 voxels on z number of points input: 1 number of voxels modified: 3690 $v122: size 100 X 100 X 100 vxrange [-4:4] vyrange[-4:4] vzrange[-4:4] non-zero voxel values: min 1.1e-14 max 97 mean 25 stddev 19 number of zero voxels: 992104 (99.21%) $v032: size 100 X 100 X 100 vxrange [-4:4] vyrange[-4:4] vzrange[-4:4] non-zero voxel values: min 0.061 max 95 mean 25 stddev 19 number of zero voxels: 992070 (99.21%) $v201: (active) size 25 X 25 X 25 vxrange [0:5] vyrange[-2.5:2.5] vzrange[-1:4] non-zero voxel values: min 15 max 97 mean 26 stddev 12 number of zero voxels: 11935 (76.38%) to continue to continue********************** file isosurface.dem ********************* to continuevfill from + : radius 0.9 gives a brick of 15 voxels on x, 15 voxels on y, 5 voxels on z Warning: voxel grid spacing on x, y, and z is very anisotropic. Consider using vgfill rather than vfill number of points input: 55 number of voxels modified: 29540 to continue to continue to continue*********************** file watchpoints.dem ********************* Plot title: plot FOO smooth cnormal watch y=.25 watch y=.50 watch y=.75 Watch 1 target y = 0.25 (1 hits) hit 1 x 50.6 y 0.25 Watch 2 target y = 0.5 (1 hits) hit 1 x 63.6 y 0.5 Watch 3 target y = 0.75 (1 hits) hit 1 x 68.3 y 0.75 to continue to continue Variables beginning with INTERSECT: INTERSECT_X = 167.511137525825 INTERSECT_Y = 20.2444312370877 to continue"watchpoints.dem" line 81: undefined function: FresnelC make[4]: *** [Makefile:742: check-noninteractive] Error 1 make[4]: Leaving directory '$(@D)/demo' make[3]: *** [Makefile:596: check-am] Error 2 make[3]: Leaving directory '$(@D)/demo' make[2]: *** [Makefile:446: check-recursive] Error 1 make[2]: Leaving directory '$(@D)/demo' make[1]: *** [Makefile:425: check-recursive] Error 1 make[1]: Leaving directory '$(@D)'