Both compiling the core and the shaders with __float128 now work.
The support was added to be ready to deal with academic edge cases.
Performance drops to 10% of double - thus disabled by default.
__float128 is a gnu extension delivering true 128 bit floating point
support. Currently not supported by clang. In addition, the quadmath
library needs to be linked. For the time being therefore disabled.
Renamed cast member of the constants struct to magic to clearly identify
magic numbers.
Introduced nlconst struct inheriting from plib::constants<nl_fptype> to
make code better understandable.
The newly added RELTOL and VNTOL parameters implement Newton convergence
checks comparable following other SPICE implementations.
The ACCURACY solver parameter now is only used for convergence checks in
iterative solvers.
In addition, type safety was significantly improved and a lot of "magic"
numbers are identifiable now.
- Added new solver parameter FPTYPE. This determines in which floating
point domain the linear system is solved. May be one of "FLOAT",
"DOUBLE" or "LONGDOUBLE"
- Added option "--fperr" to nltool. This enables floating point
exceptions. This helps debugging the code under gdb.
The purpose of this going forward is to have more choice in
optimization. Non-dynamic systems should be just fine in the float
domain. Dynamic systems (i.e. diodes, bjts, mosfets) should in general
work with double. Certain edge cases may require long double resolution.
Added ability to compile using float instead of double. Specifically the
the solver as well as the infrastructure now can have their own floating
point type. Currently this is only an academic exercise since
numerically demanding circuits like kidniki only work with double/double
support. Using float here is pushing numerical stability over the
limits.
The long term design goal is too have the matrix type (double/float)
being a parameter.
- solver: align matrix population along the various solvers
- solver: delete dead code
- renamed nl_double to nl_fptype and use nl_fptype where previously
double has been used.
- renamed param_double_t to param_fp_t
- Removed code no longer used
- Add noexcept where appropriate
- split pparser.[c|h] into ppreprocessor and ptokenizer
- smaller optimizations, e.g. use of std::size_t
- fix lint warnings
- solver now uses dynamic allocation on systems larger than 512x512
- fixed osx build
- moved nl_lists.h classes to plists.h
- fixed netlist makefile clint section
- readability and typos
- optimized the core queue dispatching logic. Minor performance
increase.
- fixed a number of bugs in parray. Now parray<double, 0> will be purely
dynamic allocation with the number of elements passed in the
constructor.
- Added noexpr where appropriate.
- Simplified the queue
Checked with gcc-7 (ubuntu), gcc-9, clang-10, macosx clang 10, mingw
cross compile on linux.
- moved netlists out of driver code into audio/ or machine/ as
nl_xxx.cpp files.
- identified and documented extended validation
- updated arcade, mess and nl targets
- Fix SUBMODEL
- move to strongly typed matrix sort constant
- extend maximum matrix size to 512x512
- optionally do parallel processing based on total operations
- templatize GMRES solver loops
Runtime performance statistics can now be enabled with nltool
option "-s". To enable those with MAME you need to run
NL_STATS=1 ./mamenl64 -v -oslog game
- Untie diode code
- Fix some typos
- add TYPE=2 opamp model which omits output voltage limitation
Useful for determining causes of non-convergence
- Fix MB3614 opamp parameters to match datasheet
This is a significant improvement to the MOS transistor model. It adds
modelling of the Meyer capacitance model.
This is a somewhat academic addition since the effects occur on a
nanosecond time scale and have a huge impact on performance. I plan to
make the capacitance model selectable. Both on a model level as well as
by introducing a global solver parameter.
The model delivers comparable results to LTSpice.
This effectively reverts b380514764 and
c24473ddff, restoring the state at
598cd52272.
Before pushing, please check that what you're about to push is sane.
Check your local commit log and ensure there isn't anything out-of-place
before pushing to mainline. When things like this happen, it wastes
everyone's time. I really don't need this in a week when real work™ is
busting my balls and I'm behind where I want to be with preparing for
MAME release.
- added MOSFET model. Currently capacitances are not modelled.
This is a 3-pin model (Bulk connected to Source) with provisions to
extend it to 4-pin at a later stage.
- Add a capacitor generic model which is charge conserving.
Switch netlist to use this model instead of constant capacity model.
- Start putting constants into a central place.
Please expect minor timing differences due to a different numerical
path.
The cmos inverter example illustrates the analog implementation of a
cmos inverter gate. These were used a lot back in the 70s/80s to
generate sinus waves. The model should also be able to better emulate
4066 analog switches.
The addition of a relatively simple capacitor model is planned at a
later stage.
Expect everything from the MOSFET model at the current stage. Wrong
results as well as convergence issues and crashes.
Change the sign of go (or in other terms a12 and a21 matrix stencil
elements). This should make further optimization of matrix population
easier.
In addition hopefully improve the readability of the code by sacrifying
overloads for more verbose member names.
