Plutonic Rainbows

With the 2026 Munich High End show on the horizon, I find myself increasingly excited about what promises to be one of the most significant gatherings in the audio industry's calendar. After last year's remarkable comeback post-pandemic, my expectations are running high. I've been hearing whispers about breakthrough Class D amplification technologies that could finally bridge the gap between efficiency and sonic purity, and I'm particularly intrigued by reports that several heritage brands are preparing flagship turntable releases incorporating advanced materials science and precision engineering never before seen at this price point. I'm also watching the streaming sector closely — rumours of lossless audio partnerships and proprietary room correction algorithms suggest we might need to rethink how we approach digital playback in high-end systems.

What I find most intriguing are the persistent rumours of collaborations between traditional hi-fi manufacturers and companies from adjacent industries — I'm imagining aerospace-grade materials in speaker cabinets and automotive engineering principles applied to vibration control. The Munich show has always been a bellwether for where the industry is heading, and I suspect 2026 will challenge my assumptions about what constitutes state-of-the-art audio reproduction. Whether these predictions materialise or the show surprises me with entirely unexpected innovations, one thing I'm certain of: Munich in May will once again be the place where the future of high-fidelity audio reveals itself to the world.

I've been eager to explore Francis Kurkdjian's latest addition to La Collection Privée, which takes its inspiration from one of Dior's most iconic fashion pieces: the Saddle bag, originally designed by John Galliano in the early 2000s. What strikes me immediately is how Cuir Saddle deliberately subverts my expectations of what a leather fragrance should be. Rather than pursuing the heavy, smoky, tar-like qualities that define traditional leather compositions, Kurkdjian has softened and relaxed the structure into something closer to supple suede. I find the result fascinating — it blends conventional pyrogenated leather notes with a musky, creamy floral cloud and modern ambered woods, creating a scent that feels both tenacious and unexpectedly light, sensual yet approachable.

I should note this isn't Dior's first attempt at capturing the essence of fine leather goods in a bottle — I remember François Demachy's discontinued Cuir Cannage from 2014, which explored similar territory. But Cuir Saddle feels to me like a more refined evolution, one that acknowledges the codes of classic leather perfumery while steering them in a contemporary direction. Available as an Eau de Parfum starting at $220, it arrives in the signature cylindrical Collection Privée flacon accompanied by an elegant leather sheath that echoes the design language of its namesake bag. For those of us who find traditional leather fragrances too assertive, this softer interpretation might be precisely what I've been missing from the genre.

Reading reviews on the Extrait De Parfum editions that were released a few months ago.

So it turns out that I have to retrain images using the Flux.2 training model — this is in order to create loras that can be used with Flux.2 [dev].

This album reminds me of the comments once made by the film director, Bobby Roth on using Tangerine Dream to score a movie about golf.

Bobby Roth in 1991:

I'd done two films with TD prior to Dead Solid Perfect and I've done two more since. I chose them for a 'Dan Jenkins' working-class comedy about golf precisely because of how strange it seemed.

I tried both American blues and country and western songs with the rough cut but both seemed too much in keeping with the existing images. Neither lifted the film to a new place or transformed it with counterpoint. Choosing a German band known for electronic music seemed bizarre to the film's producers, but when they heard the score they loved it as I did. It made golf more interesting and let people who were adverse to the game see it with new eyes. Ask any hard and fast golfer about the film and they'll say the music is perfect (though it's certainly not music one would associate with the game under traditional circumstances). Ironically, of the five films I've done with Tangerine Dream, [...] the score for Dead Solid Perfect probably did the most for the movie.

Boomkat:

OPN's anachronistic soundtrack to Josh Safdie's Timothée Chalamet vehicle is a bedazzled synth-heavy celebration of influences he's been wearing on his sleeve since day one: Tangerine Dream, Jean-Michel Jarre, Ryuichi Sakamoto, Peter Gabriel and Philip Glass.

You might wonder why Lopatin decided to go back to the 1980s once again for this one when Safdie's film, a fictional biopic of Marty Reisman, is set in the 1950s. Well, Safdie was fascinated by table tennis as a kid in the 1980s and when he was brainstorming with Lopatin in the early stages of the process, was batting across names like New Order, Tears for Fears and Constance Demby - clearly inspirations that Lopatin was equally familiar with. So the project plays like an homage to their own teenage years, with nods to classic '80s film soundtracks and enough synthesized and sampled mallet sounds (that's gotta be a Fairlight CMI, right?) to make the ping-pong theme stand out whether you've peeped the movie itself or not.

Fans of Lopatin's earliest Oneohtrix Point Never records will be stoked; it's 'Russian Mind' upgraded, in many ways, with some of the composer's loftier influences (we can hear Morricone, Jarre, Vangelis and Glass quite clearly on the ambitious standout 'Holocaust Honey') realized properly now with a fitting budget. And although the 'Marty Supreme' soundtrack is confection next to this year's mind-altering 'Tranquilizer' set, it's undeniably enjoyable. If you enjoyed Lopatin's cues for 'Good Time' and 'Uncut Gems', this just takes it even further, proving that he's one of the most capable artists in the Hollywood ecosystem right now. If the 'Stranger Things' soundtrack has gone down in history as the Demogorgon of forced nostalgia, 'Marty Supreme' is fighting the good fight.

Buy It Here

Mostly have all the lights up now. If I had packed them away properly last year, I could have got things done in half the time.

Spent some time converting DSF files to FLAC to make listening a bit easier.

Photographed for Marie Claire Japan, April 1987.

Image

I appeared as the lead in Serious Money by Caryl Churchill at the Bowen West Theatre in Bedford on the evenings of 29 and 30 November 1990. At the time, it felt immediate rather than significant. Rehearsals, performances, conversations in corridors and bars afterwards — it was all lived in the present tense. What I could not have known was that this was a pre-internet moment, one of the last times in my life when experiences were allowed to happen fully and then disappear without trace.

The play itself was only part of what occurred. During rehearsals and performances I met many girls — not in any dramatic or cinematic sense, but in the ordinary, charged way that proximity creates. Faces, gestures, brief intimacies, conversations that went nowhere but still mattered. None of this was recorded. None of it circulated. When it ended, it ended completely. What remains now are faces without names, impressions without continuity. Recognition without access. That kind of memory does not fade; it lingers, unresolved.

Over time, the memory of those nights has grown heavier than the original experience ever was. Not because the performances were exceptional, but because they have come to carry far more than they were meant to. The Bowen West Theatre has since been demolished and replaced with residential flats. The physical space that once held those evenings no longer exists. There is no digital residue to soften the loss — no footage, no archive, no searchable proof that it happened. The memory exists entirely outside technology, and because of that it feels both vivid and unstable.

This is how a memory comes to outweigh its original occurrence. It absorbs the disappearance of place, the loss of social density, and the knowledge that the conditions that produced it cannot be recreated. The memory acquires a kind of autonomy. It no longer belongs to November 1990 alone; it intrudes into the present, shaping how later life is perceived. What followed feels thinner by comparison, more constrained. In that sense, the memory has not merely survived — it has come to delineate, and at times debilitate, my life.

There is also something particularly haunting about remembering people rather than events. Buildings can be demolished and named as lost. Years can be closed off. But people vanish quietly. Those faces remain suspended in time, untouched by aging or outcome, standing in for a moment when connection felt abundant and unforced. They represent not relationships that ended, but possibilities that never had the chance to become anything at all.

In a post-internet world, moments rarely end. They persist as images, fragments, and references, endlessly retrievable. But this did not. It belonged to a world that assumed finitude — that allowed things to happen, matter deeply, and then disappear. That is what gives it its weight now. Time has moved on without hesitation, but the memory remains disproportionate, heavy not because it was perfect, but because it was fully lived and unrecoverable.

OpenAI:

Today we’re releasing GPT‑5.2-Codex, the most advanced agentic coding model yet for complex, real-world software engineering. GPT‑5.2-Codex is a version of GPT‑5.2 further optimised for agentic coding in Codex, including improvements on long-horizon work through context compaction, stronger performance on large code changes like refactors and migrations, improved performance in Windows environments, and significantly stronger cybersecurity capabilities.

GPT‑5.2-Codex builds on GPT‑5.2’s strengths in professional knowledge work and GPT‑5.1-Codex-Max's frontier agentic coding and terminal-using capabilities. GPT‑5.2-Codex is now better at long-context understanding, reliable tool calling, improved factuality, and native compaction, making it a more dependable partner for long running coding tasks, while remaining token-efficient in its reasoning.