Activity for J-F Paquet
-
J-F Paquet posted a comment on discussion Hydrodynamics - theory
Dear Sanchari, Yes, "Fvn_y-211_pT_0.01_3.dat" would be the p-T integrated flow coefficients of \pi^+ (211), integrated from p_T=0.01 GeV to 3 GeV, as a function of rapidity "y", after hadronic decays of unstable resonances are performed. Best, J-F
-
Hi Chun, The new version of MUSIC does not include Glauber initial condition, though, right? What would be the best place for Sanchari to get an MC-Glauber + MUSIC wrapper? Many thanks, J-F
-
I got a private question through my Sourceforge email that I can't seem to be able to reply to. I won't quote the question here since it was sent to me in private, but let me post the answer here for posterity. The question was about the numerical stability of MUSIC, in particular about the numerical accuracy of MUSIC's solution decreasing over time. The first thing to try would be to play with the more run-of-the-mill numerical parameters like the grid size (Grid_size_in_x , Grid_size_in_y) and...
-
Dear Farid, Yes, the flow velocity correspond to (u^tau,u^x,u^y,u^eta) Best, J-F
-
Dear Nikhil, The new version of MUSIC (the one on Github) does not currently include Glauber initial conditions. We may add it back eventually but we haven't done so yet. You could try using IP-Glasma initial conditions, as Chun discussed here: https://sourceforge.net/p/music-hydro/discussion/issues/thread/efde9a81df/#cb6f/eb39 Best, Jean-Francois
-
Hello, You are right, the temperature returned by "get_temperature(\epsilon, 0.0)" in fermi^-1, my reply was ambiguous. (I fixed it for posterity) Could it be that you are getting incorrect outputs because you are trying to output integers (ix,iy,ieta) as doubles (%e)? There is an example in function output_evolution_for_movie() of how to output the position of each grid point, for the new Github version of MUSIC. This is for x, you can see the equivalent for y and eta a few lines above: https://github.com/MUSIC-fluid/MUSIC/blob/public_stable/src/grid_info.cpp#L931...
-
The MUSIC version on GitHub is the latest version. It is faster and uses less memory. It is parallelized with OpenMP rather than MPI. We are slowly transitioning to this new version. There is nothing wrong with the version on Sourceforge however: it can still be used without problem. The new Github version of MUSIC saves less information than the Sourceforge version. Essentially, it saves only the energy-momentum tensor, in the form of the energy density \epsilon, the flow velocity u^\mu, the bulk...
-
The MUSIC version on GitHub is the latest version. It is faster and uses less memory. It is parallelized with OpenMP rather than MPI. We are slowly transitioning to this new version. There is nothing wrong with the version on Sourceforge however: it can still be used without problem. The new Github version of MUSIC saves less information than the Sourceforge version. Essentially, it saves only the energy-momentum tensor, in the form of the energy density \epsilon, the flow velocity u^\mu, the bulk...
-
The MUSIC version on GitHub is the latest version. It is faster and uses less memory. It is parallelized with OpenMP rather than MPI. We are slowly transitioning to this new version. There is nothing wrong with the version on Sourceforge however: it can still be used without problem. The new Github version of MUSIC saves less information than the Sourceforge version. Essentially, it saves only the energy-momentum tensor, in the form of the energy density \epsilon, the flow velocity u^\mu, the bulk...
-
Dear Honghao, Thank you for writing back with information on what was happening. I expect other MUSIC users will run into a similar problem some day, and will benefit from the information. All the best, J-F
-
Dear Honghao, This seems to be a generic linking problem. I think you are right that your best bet is to write a simple "example code" to better understand how to link to GSL properly on your server. If you still have trouble linking MUSIC to GSL even after you successfully link GSL to other codes, let us know: that would mean there is a MUSIC-specific problem. Best, J-F
-
Dear Honghao, Are you saying that you can compile MUSIC on your on computer? But it does not compile on another computer (the "server"), and the problem seems to be GSL? My understanding is that the "-L" argument is not always necessary. For example, I would think it is not necessary if the library is installed in one of the standard directories where libraries are installed. Also, which version of MUSIC are you using? Best, Jean-Francois
-
My apologies for the late reply. The function "Grid_info::OutputEvolutionDataXYEta()" can already outputs the temperature for the entire discretized hydro grids (tau, x, y and eta): https://sourceforge.net/p/music-hydro/code/ci/public_release_prep/tree/src/grid_info.cpp#l19 There is additional information here: https://webhome.phy.duke.edu/~jp401/music_manual/additional.html There is a function called "get_temperature(epsilon, muB)" that will return the temperature from the energy density. You can...
-
Hi Mayank, I have not been running on multiple cores for some time now, but this certainly should not happen. I looked into the code and there does seem to be a "rank==0" condition. Not in evolution.cpp actually, but in grid_info.cpp: https://sourceforge.net/p/music-hydro/code/ci/public_release_prep/tree/src/grid_info.cpp#l122 Is this the version of the code you are using? Let me know. Best, J-F
-
Welcome to the MUSIC discussion section. Should you have any question about MUSIC that is not answered in the documentation included with the sourcecode, or should you have any issue running the code, please feel free to post any question in the most relevant sub-section of the Discussion section. We expect many recurrent questions, and as such, you may want to search the existing posts of the discussion section before posting your own question (you can search using the search bar on the left, between...
-
Thank you for the status update. We haven't been able to identify the issue you were encountering, but at least you can run the code now on your updated system. Best, J-F
-
Dear Zhidong, Mode 3 and 4 use a completely different part of the code. The problem you are encountering seems to be related to a specific part to the hydro evolution (mode 2). We've run MUSIC without encountering this problem at least on OS X and Fedora ~27 (on Fedora, with gcc 7.3.1). I'm pretty sure we used it elsewhere since it's public release, but I don't remember off the top of my head. What is troubling is that, as far as I know, the only reasons a code produces different results on different...
-
Hi Zhidong, We're a little stumped. Do you compile the code with CMake or with the makefile? Whichever you use, could you reduce optimization level (in the makefile, you can just replace -O3 by -O1; should be similar with CMake), delete mpihydro, recompile everything with "-jB", run again, and post the output that you get? You should get the same results with different levels of optimization. But if you don't, then we have some clues to what is going on. Best, J-F
-
Could you post the output of MUSIC? You can just copy and paste everything that MUSIC outputs if that is easier than sending a screent shot. Thank you
-
Let's see if we can isolate the problem. Let's focus on the input file "examples_inputs/2D/music_input_mode_2", which is a Monte Carlo Glauber event. We first need to remove the randomness of this initial condition so that we can both run exactly the same thing. This can be done by fixing the random generator seed. Go in the source file "init.cpp", and search for "rnum" (line 807). Change: "double rnum = time(0) + DATA->seed;" for "double rnum = 2;" Delete the old "mpihydro" executable everywhere...
-
Just to make sure, did you modify the example input files before running with them? If not, can I ask if you are using the very latest version of MUSIC? I just downloaded a clean copy of the code and ran "mpirun -np 1 ./mpihydro examples_inputs/2D/music_input_mode_2" without any problem. Same for "./mpihydro tests/test_Glauber/music_input_mode_2 " so I'm not sure what might be happening here.
-
Dear Sruthy, Roughly speaking, the first part of a MUSIC simulation is hydrodynamics, the second part is particlization into hadrons, and the third part is hadronic decays. The second step, "particlization", is performed using the particlization hypersurface, which is saved into a file called "surface.dat". You can find information on particlization here: https://webhome.phy.duke.edu/~jp401/music_manual/hydro.html#construction-of-the-particlization-hypersurface and here https://webhome.phy.duke.edu/~jp401/music_manual/particlization.html...
-
Dear Sruthy, I don't believe Trento initial conditions can be read directly by MUSIC, but it should be straightforward to find a way to read them in. I explained here how to read external initial condition files into MUSIC: https://webhome.phy.duke.edu/~jp401/music_manual/initial.html#reading-in-external-hydrodynamic-initial-conditions The neatest option is probably to modify this function to be able to read Trento's output directly. Another option is to modify Trento's output such that it matches...
-
Dear Sruthy, I don't belive Trento initial conditions can be read directly by MUSIC, but it should be straightforward to find a way to read them in. I explained here how to read external initial condition files into MUSIC: https://webhome.phy.duke.edu/~jp401/music_manual/initial.html#reading-in-external-hydrodynamic-initial-conditions The neatest option is probably to modify this function to be able to read Trento's output directly. Another option is to modify Trento's output such that it matches...
-
Home
-
Additional documentation on MUSIC is now available here: https://webhome.phy.duke.edu/~jp401/music_manual/ The code is discussed alongside with the physics implemented in the code. It should be a very good starting point if you want to learn how to use MUSIC.
-
File "NuPInSP.dat" is a temporary file created when Glauber initial conditions are used. File "surface.dat" contains the information on the freeze-out hypersurface. As you stated correctly, the format of the file is discussed in the notes accompanying the code, in Section 1.3 (Subsection "mode 2") Files "evolution.dat" and "OSCAR.dat" are presumably empty. It is one of many files that used to be created but that we tried to clean up before making MUSIC available online. We'll look into fixing this....
-
File "NuPInSP.dat" is a temporary file created when Glauber initial conditions are used. File "surface.dat" contains the information on the freeze-out hypersurface. As you stated correctly, the format of the file is discussed in the notes accompanying the code, in Section 1.3 (Subsection "mode 2") Files "evolution.dat" and "OSCAR.dat" are presumably empty. It is one of many files that used to be created but that we tried to clean up before making MUSIC available online. We'll look into fixing this....
-
File "NuPInSP.dat" is a temporary file created when Glauber initial conditions are used. File "surface.dat" contains the information on the freeze-out hypersurface. As you stated correctly, the format of the file is discussed in the notes accompanying the code, in Section 1.3 (Subsection "mode 2") Files "evolution.dat" and "OSCAR.dat" are presumably empty. It is one of many files that used to be created but that we tried to clean up before making MUSIC available online. We'll look into fixing this....
-
Is the warning really appearing after "[Info] 0.6623 MB reading particle 2: 111 Pion(0) ..." or did you trim the log file? That would mean that MUSIC is trying to compute hadronic observables after loading only the pi+ and pi0 momentum spectra, which is bound to fail. Which mode are you running? Still mode 3? J-F
-
Hi guys, Just to be clear, are you trying to run Cooper-Frye with a single CPU, or with multiple CPUs? Thanks, J-F
-
Hi Aleksas, Sorry for the slow reply, Sourceforge was having issues over the last few days. The implementation of MC Glauber used in MUSIC was described in Section IV of https://arxiv.org/pdf/1109.6289.pdf . Let me highlight these differences with some other implementations of MC-Glauber: - There is no minimum distance between nucleon sampled from the nucleus - Negative binomial fluctuations are not included in the code, that is all nucleon-nucleon collisions deposit the same amount of energy/entropy...
-
Hi Aleksas, "Initial_profile 8" is good, this reads in an external file "Initial_Distribution_input_filename ./inputs/gs/project/cqn-654-ab/jfpaquet/init_cond/rhic_jul2014/RHIC00-05-t04/1/epsilon-u-Hydro.dat" may not work. Can you put the full directory of the initial condition file, instead of the relative directory? The grid size doesn't need to be specified, it's read from the initial condition files (whatever is in the input file is overwritten). The public version of MUSIC is currently set-up...
-
Initial conditions
-
Initial conditions
-
Home
-
Hi Aleksas, "Initial_profile 8" is good, this reads in an external file "Initial_Distribution_input_filename ./inputs/gs/project/cqn-654-ab/jfpaquet/init_cond/rhic_jul2014/RHIC00-05-t04/1/epsilon-u-Hydro.dat" may not work. Can you put the full directory of the initial condition file, instead of the relative directory? The grid size doesn't need to be specified, it's read from the initial condition files (whatever is in the input file is overwritten). The public version of MUSIC is currently set-up...
-
Welcome to the MUSIC discussion section. Should you have any question about MUSIC that is not answered in the documentation included with the sourcecode, or should you have any issue running the code, please feel free to post any question in the most relevant sub-section of this discussion section. We expect many recurrent questions, and as such, you may want to search the existing posts of the discussion section before posting your own question (you can search using the search bar on the left, between...
-
Get the speed of sound the old-fashioned way, which for currently unknown reasons appear to make the code more stable
-
Should you make changes to MUSIC that you believe would benefit from being shared with the wider community, please contact the code’s maintainers, through this website if possible, or by email. Contributions will be considered whenever possible. While no promises can be made regarding the inclusion of these modifications in the main version of the code, they could be made available separately on the MUSIC website for other users’ convenience. Bug fixes will be considered separately and will be applied...
-
MUSIC users may need to modify various parts of the code. The GPL licence allows for such modifications to be made and distributed (subject to restrictions that can be found in the licence). Nevertheless, we ask and require that you state clearly any scientifically relevant modifications made to the code when you share this modified version of MUSIC. This is to avoid any misunderstanding about the physics behind the code.
-
MUSIC is a 3+1D relativistic fluid dynamics simulation that can model the evolution of the plasma formed in heavy ion collisions, compute the thermal spectra of hadrons using Cooper-Frye, evaluate the decay of unstable hadrons, and compute various hadronic observables with the thermal or post-decay spectra. New results computed with MUSIC may necessitate modifying certain parts of the code (routines computing hadronic observables, transport coefficients, initial conditions, ...). To avoid any confusion...
-
Revert "Modified a few names in "pdg-urqmd_v3.3+.dat" so that MUSIC can read the file properly"
-
Modified a few names in "pdg-urqmd_v3.3+.dat" so that MUSIC can read the file properly
-
Other fix for recent modification related environment variable. Everything should work now
-
Fixed issues introduced with recent streamlining of eos.cpp
-
Added comments and changed default value for in...
-
Small fixes to pre-initialisation code
-
Added comments and warnings
-
Make sure mode 13/14 are run along with modes 1...
-
Fix to eta profile enveloppe when boost-invaria...
-
Small changes to makefiles
-
Small change to parameter read-in for backward ...
-
Numerical parameters
-
Numerical parameters
-
Numerical parameters
-
Numerical parameters
-
Hi Matthew, Could you provide us with a summary of the physics problem you are trying...
-
The following issue was reported by Matt Heffernan: run with "mpirun -np 2 ./mpihydro...
-
Welcome to the MUSIC discussion section. Should you have any question about MUSIC...
-
See https://arxiv.org/pdf/1202.4551.pdf for a derivation of the second order hydro...
-
Good evening, I was wondering which second order terms are currently in MUSIC, and...
1
