zernike DM vs stackarray DM

[UlasGokay]

Dear all,
i am using a YAO v5.4.0 on Yorick v2.2.04 with ubuntu 64 emulated in a virtual machine on my windows system,
I found that the zernike DM is yielding to SR of very close to 0 while stackarray DM is yielding to SR of 0.7 for the same hartmann type WFS with the same parameters. I am pasting contents of my .par file below, any comments?
//-------------------------------
sim.name = "SH_trial1";
sim.pupildiam = 128; //256
sim.debug = 0;
sim.verbose = 0;
//-------------------------------
atm.dr0at05mic = 42.44;
atm.screen = &("/home/usg/Desktop/Ulas_Yao_Works/"+"screen4096_"+["1","2","3","4","5","6","7","8","9"]+".fits"); //
atm.layerfrac = &([0.5328,0.0145,0.035,0.0957,0.1083,0.0437,0.0658,0.0371,0.0671]);
atm.layerspeed = &([15,13,13,9,9,15,25,40,21]);
atm.layeralt = &([48.,162,324,649,1299,2598,5196,10392,20785]);
atm.winddir = &([0,10,340,30,300,290,280,90,0]);

//-------------------------------
tel.diam = 3.94;
tel.cobs = 0.247462;

//-------------------------------
nwfs = 1;
wfs = array(wfss,nwfs);

n = 1;
wfs(n).type = "hartmann";
wfs(n).lambda = 0.50;
wfs(n).gspos = [0.,0.];
wfs(n).gsalt = 0.;
wfs(n).gsmag = 1.; //
wfs(n).shmethod = 2;

wfs(n).shnxsub = 10; //4 8 10 12 16
wfs(n).npixpersub = sim.pupildiam/wfs(n).shnxsub;
wfs(n).pixsize = 0.15; //0.02127; //0.15; //0.40;
wfs(n).npixels = 128;

wfs(n).noise = 1;
wfs(n).ron = 1.0; //2.0, 3.0
wfs(n).shthreshold = 0.;
wfs(n).nintegcycles= 1;
wfs(n).pupoffset = 0.;
wfs(n).fstop = "round";

//-------------------------------
ndm = 2;
dm = array(dms,ndm);

//n =1;
//dm(n).type = "zernike";
//dm(n).nzer = 80;
//dm(n).alt = 0.0;
//dm(n).unitpervolt = 0.01; //0.01;
//dm(n).push4imat = 100;
//dm(n).regparam = 0.1;
//dm(n).nxact = wfs(n).shnxsub+1;
/dm(n).pitch = wfs(n).npixpersub;

n =1;
dm(n).type = "stackarray";
dm(n).iffile = "";
dm(n).nxact = wfs(n).shnxsub+1;
dm(n).pitch = wfs(n).npixpersub;
dm(n).alt = 0.;
dm(n).unitpervolt = 0.01;
dm(n).push4imat = 100;

n =2;
dm(n).type = "tiptilt";
dm(n).iffile = "";
dm(n).alt = 0.;
dm(n).unitpervolt = 0.0005;
dm(n).push4imat = 400;
//-------------------------------
mat.method = "mmse";
//-------------------------------
target.lambda = &([1.65]);
target.xposition = &([0.]);
target.yposition = &([0]);
target.dispzoom = &([1.]);
//-------------------------------
gs.zeropoint = 1.5e11;
//-------------------------------
loop.gain = 0.6;
loop.framedelay = 1;
loop.niter = 1000; //5000
loop.ittime = 2e-3; //2e-3
loop.startskip = 10;
loop.skipevery = 10000;
loop.skipby = 10000;
loop.modalgainfile = "simulModeGains.fits";
//-------------------------------
Kind regards,
Ulaş

[frigaut]

Dear Ulas, Sorry for the lack of response ! :-( I have been mostly off during the Xmas holidays and now I am trying to keep up with a long list of emails. I will try to answer your email tomorrow. Francois

…

On 8 Jan 2018, at 11:34 pm, UlasGokay ***@***.***> wrote: Dear all, i am using a YAO v5.4.0 on Yorick v2.2.04 with ubuntu 64 emulated in a virtual machine on my windows system, I found that the zernike DM is yielding to SR of very close to 0 while stackarray DM is yielding to SR of 0.7 for the same hartmann type WFS with the same parameters. I am pasting contents of my .par file below, any comments? //------------------------------- sim.name = "SH_trial1"; sim.pupildiam = 128; //256 sim.debug = 0; sim.verbose = 0; //------------------------------- atm.dr0at05mic = 42.44; atm.screen = &("/home/usg/Desktop/Ulas_Yao_Works/"+"screen4096_"+["1","2","3","4","5","6","7","8","9"]+".fits"); // atm.layerfrac = &([0.5328,0.0145,0.035,0.0957,0.1083,0.0437,0.0658,0.0371,0.0671]); atm.layerspeed = &([15,13,13,9,9,15,25,40,21]); atm.layeralt = &([48.,162,324,649,1299,2598,5196,10392,20785]); atm.winddir = &([0,10,340,30,300,290,280,90,0]); //------------------------------- tel.diam = 3.94; tel.cobs = 0.247462; //------------------------------- nwfs = 1; wfs = array(wfss,nwfs); n = 1; wfs(n).type = "hartmann"; wfs(n).lambda = 0.50; wfs(n).gspos = [0.,0.]; wfs(n).gsalt = 0.; wfs(n).gsmag = 1.; // wfs(n).shmethod = 2; wfs(n).shnxsub = 10; //4 8 10 12 16 wfs(n).npixpersub = sim.pupildiam/wfs(n).shnxsub; wfs(n).pixsize = 0.15; //0.02127; //0.15; //0.40; wfs(n).npixels = 128; wfs(n).noise = 1; wfs(n).ron = 1.0; //2.0, 3.0 wfs(n).shthreshold = 0.; wfs(n).nintegcycles= 1; wfs(n).pupoffset = 0.; wfs(n).fstop = "round"; //------------------------------- ndm = 2; dm = array(dms,ndm); //n =1; //dm(n).type = "zernike"; //dm(n).nzer = 80; //dm(n).alt = 0.0; //dm(n).unitpervolt = 0.01; //0.01; //dm(n).push4imat = 100; //dm(n).regparam = 0.1; //dm(n).nxact = wfs(n).shnxsub+1; /dm(n).pitch = wfs(n).npixpersub; n =1; dm(n).type = "stackarray"; dm(n).iffile = ""; dm(n).nxact = wfs(n).shnxsub+1; dm(n).pitch = wfs(n).npixpersub; dm(n).alt = 0.; dm(n).unitpervolt = 0.01; dm(n).push4imat = 100; n =2; dm(n).type = "tiptilt"; dm(n).iffile = ""; dm(n).alt = 0.; dm(n).unitpervolt = 0.0005; dm(n).push4imat = 400; //------------------------------- mat.method = "mmse"; //------------------------------- target.lambda = &([1.65]); target.xposition = &([0.]); target.yposition = &([0]); target.dispzoom = &([1.]); //------------------------------- gs.zeropoint = 1.5e11; //------------------------------- loop.gain = 0.6; loop.framedelay = 1; loop.niter = 1000; //5000 loop.ittime = 2e-3; //2e-3 loop.startskip = 10; loop.skipevery = 10000; loop.skipby = 10000; loop.modalgainfile = "simulModeGains.fits"; //------------------------------- Kind regards, Ulaş — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#14>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAlb4j6eR_fMiYXfyvTy4LjhfY2Yqe7xks5tIgtEgaJpZM4RWTs3>.

[UlasGokay]

Dear Francois,

I am grateful for communicating with you! I know how the work overwhelms right after holidays...

What i am trying to do is to get a performance analysis for both Shack Hartmann and Pyramid type WFS by using YAO with similar parameters for DAG, and proceed from there to make (a simple?) analysis for sky coverage. I know that PyWFS gives higher Strehl and better performance for dim stars, so i want to see it happening in simulations,

thank you very much, and kind regards

Ulaş

[UlasGokay]

Dear all,

I also have another issue: when i run PyWFS with stack array DM, i end up with the following:

dm(1).iffile set to ulas_pyr-if1.fits
dm(2).iffile set to ulas_pyr-if2.fits
ERROR (go) marked indices not same length in matrix multiply
LINE: 3863 FILE: /usr/lib/yorick/i/yao.i

I am trying to understand that line of the yao.i code, i think my issue is related to dm nact. My .par file is as follows:
//-------------------------------
sim.name = "ulas_pyr"
sim.pupildiam = 128; //256
sim.debug = 0;
sim.verbose = 0;
//-------------------------------
atm.dr0at05mic = 42.44;
atm.screen = &("/home/usg/Desktop/Ulas_Yao_Works/"+"screen4096_"+["1","2","3","4","5","6","7","8","9"]+".fits"); //
atm.layerfrac = &([0.5328,0.0145,0.035,0.0957,0.1083,0.0437,0.0658,0.0371,0.0671]);
atm.layerspeed = &([15,13,13,9,9,15,25,40,21]);
atm.layeralt = &([48.,162,324,649,1299,2598,5196,10392,20785]);
atm.winddir = &([0,10,340,30,300,290,280,90,0]);

//-------------------------------
tel.diam = 3.94;
tel.cobs = 0.247462;

//-------------------------------
nwfs = 1;
wfs = array(wfss,nwfs);

n = 1;
wfs(n).type = "pyramid";
wfs(n).fstop = "round";
wfs(n).fssize = 3.1414; //1.728; //2.764; //1.6;

wfs(n).lambda = 0.50;
wfs(n).shnxsub = 32; //10;
wfs(n).npixpersub = sim.pupildiam/wfs(n).shnxsub;

wfs(n).gspos = [0.,0.];
wfs(n).gsalt = 0.;
//wfs(n).gsmag = 5;
wfs(n).skymag = 15.;

wfs(n).noise = 1;
wfs(n).ron = 1.0;
wfs(n).darkcurrent = 0.;

wfs(n).pyr_mod_npts= 16;
wfs(n).pyr_mod_ampl= 0.45; //0.45
wfs(n).pyr_padding = 2;
wfs(n).pyr_mod_loc = "after";

//-------------------------------
ndm = 2;
dm = array(dms,ndm);
//n= 1;
//dm(n).type = "zernike"; //??
//dm(n).nzer = 80; // due to calculation in FDR
//dm(n).alt = 0.0;
//dm(n).unitpervolt = 0.01;
//dm(n).push4imat = 100;
//dm(n).regparam = 0.1;
//dm(n).nxact = wfs(n).shnxsub+1;
//dm(n).pitch = wfs(n).shnxsub; //wfs(n).npixpersub;
//dm(n).gain = 1.;

n = 1;
dm(n).type = "stackarray";
dm(n).iffile = "";

dm(n).nxact = wfs(n).shnxsub+1;
dm(n).pitch = wfs(n).npixpersub;

dm(n).alt = 0.;
dm(n).unitpervolt = 1.0;
dm(n).push4imat = 0.1;
dm(n).elt = 1;
dm(n).gain = 1.;
dm(n).regtype = "laplacian";

if (wfs(1).type == "hartmann"){
dm(1).regparam = 1e-2;
} else {dm(1).regparam = 1e-2;}
//dm(1).regparam = 3e-2;

n =2;
dm(n).type = "tiptilt";
dm(n).iffile = "";
dm(n).alt = 0.;
dm(n).unitpervolt = 0.0005;
dm(n).push4imat = 400;

//-------------------------------
mat.method = "mmse";
//mat.method = "svd";
//mat.condition = &([10.]);
//mat.file = "";

//-------------------------------
target.lambda = &([1.65]);
target.xposition = &([0.]);
target.yposition = &([0]);
target.dispzoom = &([1.]);

//-------------------------------
gs.zeropoint = 1.5e11;

//-------------------------------
loop.gain = 0.6;
loop.framedelay = 1;
loop.niter = 1000; //5000
loop.startskip = 10;
loop.skipevery = 10000;
loop.skipby = 10000;
loop.ittime = 0.002;
loop.modalgainfile = "simulModeGains.fits";
// loop.leak = 0.05;

Kind regards
Ulaş