vellum
vellum's bugs are now found by using it, not testing it — the downstream grammar is doing the QA.
A side-by-side editorial comparison of mrbayes and ravetools — release velocity, themes, recent moves, and the top alternatives to consider.
mrbayes spent 2026 auditing its own Bayesian MR estimators for coding errors.
mrbayes wraps Stan and JAGS implementations of Bayesian Mendelian randomization estimators — IVW, MR-Egger, radial Egger, and their multivariable forms. Most of the visible history is packaging work: dependency trimming, conditional examples so the package installs where JAGS will not compile, a maintainer handover. The 0.5.3 release in July 2026 breaks that pattern with a dense list of fixes inside the model code itself.
RAVE's signal-processing toolkit taught base R graphics to render lit 3D brain meshes.
ravetools supplies the signal processing and now the rendering primitives behind RAVE, a suite for analyzing intracranial electroencephalography. Its earlier releases built out Matlab-equivalent filtering, Welch spectra and wavelet tooling; 0.2.6 changes register entirely, adding plot_mesh_dotcloud and plot_mesh_polygon to draw 3D surface meshes in base R graphics with no rgl dependency, plus arbitrary clipping planes and per-mesh alpha.
mrbayes wraps Stan and JAGS implementations of Bayesian Mendelian randomization estimators — IVW, MR-Egger, radial Egger, and their multivariable forms. Most of the visible history is packaging work: dependency trimming, conditional examples so the package installs where JAGS will not compile, a maintainer handover. The 0.5.3 release in July 2026 breaks that pattern with a dense list of fixes inside the model code itself.
The package has moved from packaging upkeep into a correctness-audit phase. 0.5.3 fixes a hardcoded three-exposure loop in MVMR-Egger reporting, a broken joint-prior branch, a sigma parameterization error in radial Egger, and several prior specifications — the profile of a maintainer reading their own model files closely rather than responding to bug reports. Platform work continues underneath: an R 4.3.0 floor inherited through a transitive dependency chain, and segfault fixes on macOS ARM.
Expect further audit-driven patches to the remaining rjags and Stan model files rather than new estimators; the fixes in 0.5.3 cluster in the Egger variants, which suggests that is where the reading is still in progress.
ravetools supplies the signal processing and now the rendering primitives behind RAVE, a suite for analyzing intracranial electroencephalography. Its earlier releases built out Matlab-equivalent filtering, Welch spectra and wavelet tooling; 0.2.6 changes register entirely, adding plot_mesh_dotcloud and plot_mesh_polygon to draw 3D surface meshes in base R graphics with no rgl dependency, plus arbitrary clipping planes and per-mesh alpha.
The package is absorbing the visualization layer that RAVE previously delegated to WebGL-backed renderers, and doing it in a way that works in headless and figure-generating contexts. Exporting ensure_mesh3d to coerce mesh3d, ieegio_surface, fs.surface and surf.asc into one canonical form makes ravetools the convergence point for surface formats across the wider toolchain — the sibling ieegio package shipped its own base R plot method within minutes of this release.
Expect the base R renderers to pick up the features rgl users would miss next — richer materials, labeling and camera control — and for more of the RAVE stack to route surface handling through ensure_mesh3d.
Other Analytics products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either mrbayes or ravetools.
vellum's bugs are now found by using it, not testing it — the downstream grammar is doing the QA.
n2kanalysis has spent eight years wiring INLA models to an S3 bucket.
A Fortran-descended optimizer got thread-safe, then found two flags that never worked.
ggstatsplot reached 1.0 by adding tests, having outsourced its statistics years ago.
collapse got a JSS paper and a 7x fmean speedup in the same release.
gtsummary is quietly rebuilding itself around analysis results data, one table verb at a time.
See all mrbayes alternatives → · See all ravetools alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. mrbayes and ravetools are shipping at a similar cadence (velocity 0.0 vs 0.0, both within Sparkpulse's "active" band). See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. mrbayes and ravetools are shipping at a similar cadence (velocity 0.0 vs 0.0, both within Sparkpulse's "active" band). For your specific use case, the alternatives sections above list other Analytics products to evaluate alongside.
Top mrbayes alternatives in Analytics are ranked by recent ship velocity. Browse the "mrbayes alternatives" section above for the current picks, or visit /alternatives/mrbayes for the full list with editorial commentary on each.
Top ravetools alternatives in Analytics are ranked by recent ship velocity. Browse the "ravetools alternatives" section above for the current picks, or visit /alternatives/ravetools-r for the full list with editorial commentary on each.