https://wasatchweatherweenies.blogspot.com/2019/09/how-good-is-new-gfs.html

The GFS was upgraded in June and the changes were substantial. 
The so-called "dynamical core" of the model, the guts of a numerical modeling
system that are responsible for solving the equations that govern atmospheric
motions, is all new and based on the NOAA Geophysical Fluid Dynamics Laboratory
(GFDL) Finite-Volume Cubed-Sphere dynamical core known as the FV3.

The equations that govern atmospheric motions are well known and based on
physical principles such as the conservation of mass, conservation of momentum,
ideal gas law, etc.  Solving these equations efficiently for a sphere (i.e.,
Earth) on a massively parallel computer, however, isn't straightforward.  The
old dynamical core of the GFS was dated, not optimal for modern computer
infrastructure, and problematic as model grid spacings continued to decrease. 
Thus, a change was needed.

Numerical modeling systems also need "physics" to account for things like
radiation, cloud, land-surface, and other processes that affect the atmosphere. 
These are more difficult since we either don't know or can't define the physical
equations that govern these processes, or they occur on scales that are smaller
than can be simulated directly on current computers.  Cloud processes provide a
good example.  We simply cannot simulate directly the formation of every cloud
droplet, rain drop, drizzle drop, and ice particle in a cloud.  Shortcuts must
be made.  This is called parameterization.

The new GFS is based primarily on the old GFS parameterizations, except for
replacement of the old cloud parameterization with a new one developed at GFDL
and a few tweaks to the land surface and ozone/water vapor photochemistry
parameterizations.

Overall, the results above indicate that the new GFS should perform about as
well or slightly better than the old over the western United States, with
relatively similar biases. It does not appear to perform as well as the previous
operational version of the HRES. This is especially true if one performs some
bias correction. Nevertheless, the differences are not huge, so the new GFS
warrants consideration in the forecast process and is going to outperform the
HRES in some events. Knowledge of the general lowland and mountain biases
described here should be helpful, but be aware that those biases do and will
vary on a station by station basis, which could be important for point specific
forecasts.