← /home

PRESS KIT

$ cat /press/intro.txt # for curators, journalists, and anyone writing about the project

THE90SPOPE — portrait

// not a photograph — regenerated likeness of the older online persona used across the project's accounts elsewhere.

facts.txt
artist:        THE90SPOPE
based:         Norway
origin:        Lithuania
active since:  early 2025
status:        independent · self-released
tools:         Reaper · iZotope RX · standard mastering chain
mastering:     manual · bedroom · no AI
targets:       -8 LUFS / -1 dBTP
contact:       contact form on the90spope.com

BIOS

$ ls /press/bio/ # pick a length, hit copy, paste in

bio-50.txt
THE90SPOPE — Norway-based one-person studio, Lithuanian by origin, bedroom-mastered by necessity. Tracks start as samples, hummed melodies, or beats made in Reaper. Built with AI tools, finished by hand. Active since 2025. Genre decided per track. Mastering is manual, no presets. Summoned by code, shaped by flesh.
bio-100.txt
THE90SPOPE is a one-person studio operating between the bedroom and the terminal. Norway-based, Lithuanian by origin, active under this name since early 2025 — the alias is older than that. Tracks start as samples, hummed melodies, or beats made in Reaper, then built with AI models (open-source where they cooperate, closed-source where they won't) and finished by hand. iZotope RX before the DAW, the regular mastering chain in it — multiband compressors, clippers, saturators, manual EQ — tweaked for AI sound. No AI mastering. Targets -8 LUFS / -1 dBTP when the source can take it. A track might take a week if it has to. Summoned by code, shaped by flesh.
bio-300.txt
THE90SPOPE is a one-person studio operating between the bedroom and the terminal — Norway-based, Lithuanian by origin, active under this name since early 2025. The alias predates the project; the project waited for tools that could keep up.

The producer started learning Reaper at seventeen and never switched DAWs. For a long time the music was on-and-off — more curiosity about how things sound than confidence in shipping them. AI music changed that math: instead of one idea per evening, suddenly there were infinite candidates to fix, push, and try to make sound louder and better than they had any right to.

Tracks usually begin as a hummed melody, a sampled fragment, or a beat made in Reaper. That seed is fed to AI models — Musicful AI and various closed-source generators, never Suno — to produce candidates. From there the workflow is unglamorous: iZotope RX before the DAW, with per-section gainstaging, preclipping to even out the limiter, declick, upsampling where it helps, EQ where it's missing. In the DAW it's the regular mastering chain — multiband compressors, multiband clippers, saturators, manual EQ — tweaked for AI sound. No AI mastering. Targets are -8 LUFS / -1 dBTP when the source can hold up; if it can't, the track gets pulled or reworked.

A finished track might have a week in it. Sometimes more. The output is meant to stand out from the usual AI-music output: louder where it can be, weirder where it has to be, mixed by a person who would rather spend an hour fixing a kick than ten seconds picking a preset.

Bedroom-mastered, no excuses about it. Self-described "bedroom sound engineer stage" — not a genius, just stubborn about making things sound better. Summoned by code, shaped by flesh. Use the machine without becoming one.

PRODUCTION STACK

$ cat /press/tech.txt # the chain. nothing fancy.

tech-rider.txt
[DAW]
reaper                    primary, since ~17 years old, never switched

[pre-DAW repair · iZotope RX, per fragment]
waveform gainstaging      so the limiter sees an even level
preclipping               to shave stray peaks before the limiter
declick                   when the source has digital artefacts
upsampling                where it helps
fragment EQ               where it's missing

[in the box · reaper]
multiband compressors     dynamics control across the spectrum
clippers · multiband      shape transients, hold loudness without limiter pumping
saturators · various      colour and harmonic glue
manual EQ                 surgical and tonal, no presets, no AI mastering chain

[loudness target]
-8 LUFS · -1 dBTP         when the source can hold it
                          if it can't, the track is reworked or pulled

[origin material]
hummed melody · sampled fragment · reaper beat

[ai stack]
musicful ai               and various closed-source generators
NOT suno                  on principle, the output is too uniform
treated as raw            never as finished — it always gets fixed

FEATURED RELEASES

$ ls /releases/ | head -3 # latest three; full catalog at /#music

PRESS

$ cat /press/quotes.txt

quotes.txt
// no press yet — be the first