2
[2026-09-02] · src GlueX Collaboration; Phys. Rev. Lett. (2026), arXiv:2512.04136
Using a 12 GeV photon beam on a proton target in Hall D at Jefferson Lab, the GlueX Collaboration measured the photoproduction cross section for gamma p -> phi(1020) pi+ pi- p and looked at the phi pi+ pi- invariant-mass spectrum for the established phi(2170) resonance (also called Y(2175)). The spectrum does not match the PDG-averaged phi(2170) parameters. Instead it shows a structure at m = 2239.2 +/- 13.4 (stat) MeV, width Gamma = 139.8 +/- 24.0 MeV, with significance 4.8 sigma including systematic uncertainties (5.5 sigma statistics-only) - just under the conventional 5 sigma discovery bar once systematics are folded in - and a second, weaker structure at m = 1828 +/- 12 (stat) +/- 15 (sys) MeV, Gamma = 94 +/- 33 (stat) +/- 37 (sys) MeV, at 3.6 sigma (3.7 sigma statistics-only). Neither mass nor width for the first structure agrees with the quoted phi(2170), and the second sits in a region where no established phi-family state is quoted at all.
This is evidence, not yet a discovery: the better-measured structure misses five sigma once systematics are included, and the weaker one is nowhere near it. What earns it a slot regardless is that GlueX went looking for a specific, already-catalogued resonance and got a spectrum inconsistent with that resonance's own quoted parameters - which means either the phi(2170)/Y(2175) world-average needs revising, or there are two additional states overlapping it that previous, lower-statistics measurements could not resolve. Both options say the strangeonium spectrum in this mass region is less settled than the PDG entry implies. Nothing here is a hybrid-meson confirmation or any other specific identification - the collaboration itself says it does not yet know what either structure is - so the honest read is "the phi(2170) region needs another look," not "two new particles."
[2026-09-02] · src CMS Collaboration; arXiv:2609.01385 (2026)
The Standard Model predicts a Higgs boson total width of 4.1 MeV, roughly 250 times narrower than CMS's diphoton mass resolution - a direct Breit-Wigner fit to the resonance shape cannot see it at all. CMS instead used the full Run 2 dataset (138 fb-1 at 13 TeV) and fit for the interference between resonant gg -> H -> gamma gamma production and the non-resonant gg -> gamma gamma continuum, which distorts the diphoton mass spectrum in a way that depends on the width even when the resonance itself cannot be resolved. The result is Gamma_H < 92 MeV at 95% confidence (expected limit 138 MeV) - the most stringent width bound yet from an on-shell Higgs measurement, though still about four times looser than CMS's existing 22 MeV bound from comparing off-shell to on-shell ZZ -> 4l production.
The off-shell/on-shell ratio method has held the tighter width bound for years, but it only works by assuming the Higgs couplings do not change between the off-shell and on-shell regimes - an assumption that is reasonable in the Standard Model but is exactly what several beyond-Standard-Model scenarios would violate. The diphoton interference method used here needs no such assumption: it bounds the width from on-shell data alone. That it lands at 92 MeV rather than something close to 22 MeV is not a weakness of the technique so much as a statement of how hard the on-shell problem is; what matters is that an assumption-free on-shell path to a real bound now exists and has a number attached, where before there was only the resolution-limited non-result of a direct fit. Tightening it further is a matter of more data, not a new idea.
2
[2026-09-02] · src —
Piano gets today's block alone: it carries the longest gap of the three, and alto went yesterday, so the two chops-dependent instruments don't land back to back either way. The exercise is new territory inside a family piano has already visited once — stride, eight days ago, asked for a bouncing LH alternating bass notes and chords across two octaves. Freddie Green is stride's rhythmic opposite: four flat, identical, unaccented quarter-note attacks per bar in both hands together, so still and so consistent that the chord stops sounding like harmony and starts sounding like a click track. Key is G♭ major, the most distant point on the fourths cycle from the B major this family used last time, and the most awkward key on the exercise's own list.
4 minutes: LH root alone on the G♭ ii-V (A♭m7 → D♭7, one bar each), quarter = 72, four times through — no RH yet. The point is planting the physical quarter-note pulse in one hand before asking both hands to hold it together; the exercise's own first practice step is exactly this split.
Both hands together: LH root plus RH chord stab on every beat, four attacks per bar, cycling the G♭ ii-V four times at quarter = 72. Success criterion: four consecutive bars where every attack lands at the same soft dynamic (pp) with no accidental accent landing on beat 1 — the failure this exercise is built to catch is a chord that announces itself instead of disappearing into the pulse.
Take the same pattern into a 12-bar blues in G♭, comping one full chorus with the Freddie Green rhythm under a slow blues feel. This isn't a drill at that point — it's time being delivered continuously, the way the exercise's own notes describe it: the chord becomes the click the rest of a band would lock to.
Stop the moment consistency breaks — an accent creeping onto beat 1, or the tempo starting to drift — rather than pushing through more reps once it does. The exercise depends on monotony precise enough to lock in; forcing another chorus after the ear starts noticing unevenness trains the unevenness, not the pulse.
[2026-09-02] · src —
hearHerb Ellis & Freddie Green — Rhythm Willie (Rhythm Willie) · Tidal · Spotify
The exercise being practised today is a piano translation of a guitar part that almost nobody ever actually heard clearly — on the Basie band records that made Green's name, the guitar sits so far back in the rhythm section that it registers as pulse rather than as an instrument. This 1975 quintet session, cut with Herb Ellis, Ray Brown and Jake Hanna rather than the full orchestra, is one of the rare recordings where Green is up front in the mix instead of buried under a big band, so the four flat quarter-note stabs today's technical target is asking for are actually there to hear, one at a time, against Ellis's busier lines.