(You received this message because you are a
registered user of MCX (https://mcx.space) - an open-source Monte
Carlo photon transport simulator - and had indicated that you
wanted to be notified for future updates in your registration
form. If you are no longer interested in this software, please
feel free to click on the unsubscribe link at the bottom of this
email. In that case, we apologize for the inconvenience.)
Dear MCX/MMC registered users,
I am excited to announce MCX/MMC v2025.10 - a stable release that
brings international language support (MCX), faster simulations
(MCX-CL/MMC), critical bug fixes (MMC) and across-the-board Python
support (pmcx/pmcxcl/pmmc) to our global user community. Everyone is
encouraged to update!
Please go to our website
https://mcx.space, or use the following
link to access the latest packages
https://mcx.space/wiki/?Get
I would like to highlight some of the key updates below:
=== MMC critical bug fix ===
- A critical bug related to normalization was found and fixed
for MMC, affecting MMC v2025 (released in Feb 2025). The bug
produces inaccurate normalization factor - the fewer the
cores/blocks (such as on the CPU), the less accurate. You are
urged to update your MMC/MMCLAB with this new release.
=== MCX/MCXStudio ===
- [i18n] MCX and MCX Studio add international language support,
supporting Chinese (zh_CN and zh_TW), French (fr), Spanish (es),
Germany (de), Japanese (ja), Korean (ko), Hindi (hi) and
Portuguese (pt)
- [i18n] MCX Studio Windows installer also allows users to
choose language during installation
- [pmcx] a dozen MATLAB-based MCX data processing functions (mcx2json,json2mcx,plotphotons,preview,...)
were ported to Python/pmcx, unit-tested
- [feature] creating frequency-domain Jacobians in the replay
mode - contributed by Paulliina Hirvi and Jarno Rantaharju
- [feature] multi-source photon replay
- [bug] fix mcxstudio incorrect 3D rendering of column-major
array produced after v2024.2
- [mcxstudio] the mcxstudio installer also installs iso2mesh and
redbird (a new DOT toolbox) alongside with
mcxlab/mcxlabcl/mmclab
=== MCX-CL ===
- [port] multi-source simulation ported from mcx
- [port] hyperboloid/ring/expanded slit source types ported from
mcx, new mcxlab/mcxlabcl srctype demo unit
- [port] Russian roulette and multi-source replay ported from
mcx
- [opencl] enable optimization level 3 (--optlevel 3) by
default, boosting speed by 30%-100% on various devices
- [pmcxcl] pmcxcl now supports Apple silicon
- [opencl] tested and optimized for Arm Mali GPU (Orange Pi),
accelerating by 6x
- [mcxlabcl] use USE_MCXCL>0 to specify default OpenCL device
globally
=== MMC ===
- [pmmc] brand new mmc module for Python, paired with the new
pyiso2mesh module for mesh generation
- [pmmc] pmmc supports Apple silicon
- [port] support focal-length, isotropic/Lambertian launch via
cfg.srcdir(4), like in mcx/mcxcl
- [opencl] enable optimization level 3 (--optlevel 3) by
default, boosting speed by 30%-100% on various devices
- [bug] fix immc photon-leakage bug (thanks to Aiden Lewis)
- [bug] allow immc to run on Apple silicon, previously immc was
not compiled with omp target
An animation showing installing and running MCX Studio with
international language support can be found here
https://mcx.space/tutorials/install/mcxstudio_v2025_10_installer.mp4
Windows users - you can also install the entire MCX suite
v2025.10 using the below winget command on Windows 10 or later
winget install mcxstudio
For Python users, you can directly install pmcx/pmcxcl/pmmc using
pip
python3 -m pip install pmcx pmcxcl pmmc iso2mesh
If you use docker, all mcx/mcxcl/mmc docker images
(https://hub.docker.com/r/fangqq/mcx) have also been updated, you
simply pull and run
docker pull fangqq/mcx:v2025.10
docker run --rm --gpus all "fangqq/mcx:v2025.10" mcx -L
enjoy the enhanced mcx/mmc and happy modeling!
Qianqian